Slashdot is powered by your submissions, so send in your scoop


Forgot your password?
Handhelds Hardware

Census Bureau Wants 500,000 Handhelds in 2010 124

andori writes: "ComputerWorld is reporting that the Census Bureau is wanting to conduct the 2010 Census without the use of paper. They want to use 500,000 handhelds with GPS and wireless communications abilities. And they want to do it for $100 an unit. I sure hope the industry is able to that price point some day! I will personally take a few if they do."
This discussion has been archived. No new comments can be posted.

Census Bureau Wants 500,000 Handhelds in 2010

Comments Filter:
  • We already have handhelds that can do that and by 2009ish when they will want to buy them I dont see this being an issue. We already have several handhelds for around $100 on the market.
    • Yep. All the functions they want are available now in one form or another in handhelds. In 6 years they should have them at much reduced cost. I can even see less than $100 in 500k quantities. As I see the way chips are going now. It would fit on two chips plus some support. The main CPU, memory, display and IO control will all fit on one chip by then. The GPS and cell phone part will be another chip. They may even get them merged by then, but I fell the analog side of the cell phone and GPS will likely make it so they still use two chips. The support circutry will be stuff like power supply and IO protection devices, that's it. It will likely be the size of a standard V series palm pilot or maybe even thinner. By then the back of the case will able to be the battery. Got to love the Li-Polymer batteries. Something like that might even get me to carry a cell phone.

      As a side note, they really need to ahve them by mid 2008ish. That's only 6 years off.

  • I know from firsthand exp. that would make things MUCH more simple. however it wouldn't work in alot of areas, such as mountian communities where thetre is wireless service, hell you're lucky to dialup in the mountians! but in the larger areas (where the bulk of the population is) it would make things go alot faster.
    • "however it wouldn't work in alot of areas, such as mountian communities where thetre is wireless service"

      1) we are talking 2010 here.. in 7 1/2 years a lot of things can change 2) gps uses satellites.. you don't need a cellular tower.
      • 3) GPS is one-way.... the GPS will just fix their position - the wireless link is still needed to get the data from the device back to the servers...

        To be honest at the rate of change in the last year with the handheld market it will be interesting to see what they come up with.
        • Well, if the device has enougth storage space, they could just get the data back by plugging it into its base and having it upload data to the servers from there.
  • Uhh.. (Score:4, Insightful)

    by Wakko Warner ( 324 ) on Saturday March 16, 2002 @03:12AM (#3172341) Homepage Journal
    They want to use 500,000 handhelds with GPS and wireless communications abilities. And they want to do it for $100 an unit.

    Can't the Blackberry almost do this already? In 7 years, we'll have nothing to worry about. You vastly underestimate the pace at which technological innovation moves on this planet.

    - A.P.
    • Re:Uhh.. (Score:3, Insightful)

      by KFury ( 19522 )
      Err, Blackberry can do wireless, but not GPS. A Garmin eTrek can do GPS for under $100, but not wireless.

      How necessary is wireless to the equation, really? What's wrong with storing the info on the unit until it gets to a base station at the end of the day? I mean, will wireless get the final census data out a day earlier?

      Of course, from a usability standpoint, where accuracy is absolutely vital, I don't think a chicklet keyboard like the blackberry or hiptop will cut it, and handwriting recognition probably won't either, so it'd probably be a pen-based systen with custom software. Add 24 months of development and 6 months of field testing, not to mention the specification and bidding process, and they'll have to be using technology that comes out in the next couple years...

      "You vastly underestimate the pace at which technological innovation moves on this planet."

      Perhaps, but you vastly overestimate the pace at which bureaucratic advancement moves on this planet.
      • There are two possible options: wireless or lithium rechargable.

        The problem with waiting until the PDA gets back to the cradle is that the batteries might fail, and data could be lost. Batteries that last a month between charges are one solution; wireless so the fact that the batteries might fail is irrelevant are another.

      • >Of course, from a usability standpoint, where accuracy is absolutely vital, I don't think a chicklet keyboard like the blackberry or hiptop will cut it, and handwriting recognition >probably won't either, so it'd probably be a pen-based systen with custom software.

        It would probably be a handwriting recognition system using blocky letters. That is what they used in 2000. They trained us how to write properly so that the machine wouldn't have a problem picking it up, so I don't see this being a big deal.
    • Look at the history of tech. Things get cheaper... to a point. Then they just add features instead of lowering price.
      • Sure, and at that point you can pick up some good used stuff for $100.
        • A friend of the family works for a fairly large company, and has mentioned in the past that he can get some very steep discounts on handhelds and such- ie, he was able to buy a palm m505 for $50 or so, sold to him at cost. If the census bureau buys that many handhelds, and considering the decrease in prices over time, why should there be any problem meeting the $100 price?
    • Training and implementation and testing takes time. Pencil and paper are easy. Many of these units will need to be delivered 1-2 years in advance.

      So they are right to get started on this now.

      Not to mention building the back end network.
    • Yeah, why would GPS be even necessary?

      When the people who carry out a census need a GPS to locate themselves, I'll for sure doubt about their ability to do their jobs.
      • Yeah, why would GPS be even necessary?

        That was my first thought, then it occurred to me how useful this would be. In addition to the census taker recording your address, the system records the lat & long. Not only does this help provide error correction & prevent fraudulent entries (I imagine some census takers just fill out the forms without actually making their rounds), but it also simplifies interpreting geographic data. Want to know how many people live between the 44th & 45th parallels in the US? No problem, the data's already in the system.
  • by PaxTech ( 103481 ) on Saturday March 16, 2002 @03:12AM (#3172343) Homepage
    Right now, the functionality is just about there and the price isn't that far off either. Given another 6-7 years of tech, I have to say I'll be pretty disappointed if something like that costs anywhere near $100.
    • Unless it has been banned by the SSSCA or subsequent legislation.
    • The way things are going, I'll be very disappointed if the tech isn't built in to the very people assigned to the beat 7 years down the road. mod me down! off topic, whatever. my troll is off, so an OT will hafta do. thank you.
    • Handhelds have been out for a couple of years and have made significant progress technologicaly but the prices of the low-end devices are still expensive. The manufacturers prefer to phase out old models than to have theirs high end handhelds compete with a low end model of yesterday's technology that could be sold for 50$.
    • Have you ever seen a hard disk retail new for under $100?

      Some devices have magic price points that won't be broken, even if the devices don't cost much to make. More features will be added, faster processors, better screens, etc, but I think you'll be disappointed in six years.
  • By 2010, I don't see this as a problem. Hell, the local power company is already doing this.. not for that cheap, and they're not exactly palms.. but we have 8 years to go.

    I don't see why this will be a problem.

  • Sprint PCS [] is selling the Kyocera 6035 [] for $150. The 6035 is an combination Palm and CDMA phone.

    I don't think it's too unreasonable that a combination GPS, PDA, and phone could be made available at $100/pop within the next 6 years.

  • by ubiquitin ( 28396 ) on Saturday March 16, 2002 @03:37AM (#3172363) Homepage Journal
    As the volume goes up, the price goes down. If retail stores take their typical 30+% markup on a $120 Palm device, then the Palm device makers are already in the range. I'm sure any one of the major handheld players would love to get that contract $50 million for 500,000 handhelds leaves LOTS of margin when those 500,000 handhelds are being churned out of fabs in Korea and Hong Kong.
  • Isn't the whole point of the census to get a complete count of everyone? That means, everyone needs to fill out a form... Do they expect people to do it over the web or something?
    • A couple of weeks ago it was reported that the UK government was considering abandoning censuses. They reckon that most of the information it gathers can already be got from tax, social security and other records and that the cost of the census isn't justified by the results.

      There's also a recognition that the UK population isn't as docile as it used to be and there's a suspicion that the number of people refusing to complete it is going up every ten years - the Poll Tax put off some folks back in 1991, in 2001 they didn't realise that other people would object to there not being an "English" box to tick when they asked us our ethnic origin for the first time.
      • Intresting idea, but the US Census is mandated by the Constitution, so we just can't do away with it.

        Of course, with the current occupants of the White House & Supreme Court, little things like the Constitution don't matter too much...
    • Not everyone fills out the forms, and some people don't receive them (some new houses, for example, or "off the record" divisions of houses into apartments, etc. aren't on the address database). So the Census Department follows up the form campaign with a door-to-door campaign to try to fill in the holes.
  • Heck.. (Score:1, Funny)

    by Anonymous Coward
    If they give me one of those computers, I'll actually take part in the census this time..
    • 40,000,000 million households, 500,000 units. I guess you get 1/80th of a unit. ;)

      On the other hand at $100 each, that's peanuts to buy. My guess is the unit will actually be made in million pluss quantities as it would also be quite usefull as a standard PDA/Cell Phone/GPS. I'd buy one, especially if it had a large non-volitile memory for programs and data.

  • by Yoda2 ( 522522 ) on Saturday March 16, 2002 @03:43AM (#3172374)
    Hopefully this goes without saying, but they better have a very good security model.

    Otherwise we might see small rural areas with amazing population booms, their own congressional districts, and lots of federal $$$.

  • Hopefully (Score:2, Interesting)

    by red5 ( 51324 )
    Hopefully they don't save the data in a proprietary format. Thus rendring the data useless by 2015. Like say domesday []. :)
  • Palm was selling Palm VII's for a hundred bucks for a while with large rebates. You would then need GPS units though. Still, with an order of 500k units they should be able to get a pretty deep discount so I think that $100 should be reasonable. Also, remember the next census is many years off, so prices should be lower in general (Hopefully!!).
    • in 7 years they can buy MY palm VIII (heck, make it Palm XII !) for eighty-- make it an even hundred. i'll bet my coworkers will take em up on that one too, even toss in a used GPS mod or two for that price. Or possible trade for a Big Mac. Either or.
  • by HappyCycling ( 565803 ) on Saturday March 16, 2002 @03:48AM (#3172385)
    The census people want Bejeweled, and they want it soon.

  • yet another completely absurd dependence on a machine in a system that works fine without them.

    but hey -- its more profits for some company! and we all know thats good for everyone.

    • Not exactly. Although I can't figure out why they need 500,000 to count the heads of 325 million people. Each one is good for only 625 heads? Maybe they will be sparsley used in rural towns, but I could see a single bean counter hitting the houses of 2000+ people in a weeks time in a metropolitan setting. I think 500,000 is a little overboard.

      And what about this: Where do all 500,000 go when they are done with the census? Do they keep them for the next census, or flood the market with 2 week old handhelds?

      Good questions...
      • Although I can't figure out why they need 500,000 to count the heads of 325 million people.

        I can't figure out why they would need any.

        Here we have 325m people. Count them. Your budget is $50m.

        (Yes I know censuses (censi?) aren't just for counting...)

        • Why would they need any? Well, let's see...

          You start with 325 million people.

          Let's say for the sake of argument that all of those people are members of an a household of 3. That's 108,333,333 households, round off to 108m.

          That's 108 million census forms that have to be sent out. If the form is sent with a return envelope, that's 3 paper articles each form, which effectively brings us back up to 325 million again. As I recall, I got two forms in the mail last time because I didn't send in the first one fast enough. Let's say that half of America turns in their form on the first go. That's an amazingly generous estimate, but we'll go with it.

          That's 438 million sheets of paper used up so far.

          Now we have the actual head-counters.

          Let's say that half of the people served a second census form turn it in, leaving us with just 1/4th of the population to count. That's 81m census forms. They leave you a new one every time they come by if you're not home, and I think I got 3 before they finally caught up with me. Let's say they get half the first time. Forty million are left a second, and if half of them are home, 20 million get the third. That's a total of 141 million more sheets of paper, added to our 438 of before.

          So a total of 579 million pieces of paper are used up in this venture so far. We won't bother counting the shipping materials for all those head counters to send in their finished forms.

          Now we have 108 million households worth of census info. What do we do with it? Why, stick it in a computer, of course! What good is census data if the government can't poll the database for statistics? If we employ 100 data entry operators working 8 hour shifts every day and entering an average of 1 household each every 30 seconds, it'll take us a mere 35 days to enter them all.


          We send out 325 million sheets of paper, then send out the head counters with their palmtops. You need 500,000 because half will end up broken during use. Just like you'd likely lose another 50 million sheets of paper due to carelessness on the parts of mail personnell and head counters.

          All the data collected is already on computer, so you don't need 2800 man-hours of work to key it all and there's no chance of error in the transfer from hard-copy to digital because the head of household is standing right there when you enter it.

          And, best of all, those palmtops are good for more than one use. All that paper costs money to recycle, if it even gets recycled. The PDAs that don't end up broken during use are good for the next census. Sure, they'll be a few years out of date from industry-standard, but if they're good enough for the job now, they're good enough then. So even if you Do lose half to damage, that's $25 million less you're spending on the next census, on top of money saved from printing, shipping, and data entry expenses.

          So yes, using electronic census forms is a complete and total waste of time and money. Why on earth would we ever want to do it when paper is so clearly more efficient? God knows computers are worthless. Aren't you glad we had the foresight to nip that internet thing in the bud and keep on getting Slashdot by carrier pigeon? It worked from great-grampa, it'll bloody well work for me, too.
        • The flaw in your logic is that plenty of people don't want to be counted or just can't be bothered to return the forms, etc.

          That means the Census Bureau has to send out people to knock on doors, set up appointments, and generally try to track down the people that didn't return the forms.

          Then, they also have to count homeless people and such. That's not an easy task, either.

          $50m isn't enough to count the people, at least not with any reasonable assurance of quality.

  • Sharp has a sweet looking PDA with embedded Linux but WOW, pricey! Between $500.00 and $600.00.

    I'm talking about the Zaurus SL-5500 SL-5500. But at that price I doubt that the census people will be packing them around. If you want to take a look at one you'll find it here [].

    It's amazing how much technology has changed. I wonder if they'll have a "Get the hell off my property and leave me alone!" checkbox.

    • I have a development unit of the Sharp Zaurus ... and it is sweeeeeeet.

      The user software is still being fine tuned (for consistancy and "missing" functionality, but when the consumer version is released ... watch out.

      Go over to ZaurusZone [] to check out what the open source community is doing (the entire shabang).

      Sorry I can't give you links to the developer sites, but I will say that the Trolltech contest alone will result in 1000s of available (and many free) applications.
  • What would these half-million units be used for after the 2010 census? One thing is certain: The wireless protocols they end up using won't exist in 2020. Maybe they should donate them to schools?
    • I propose that the used units are recycled into cattle feed, the way everything else is these days. Cheapest antibiotics around comes in your Quarter Pounder with Cheese :O) so why not your daily silicon and plastic too? failing that, maybe another CueCat will emerge from the wreckage, and we'll have more better toys again! Wheeeeee!
    • 500,000 *BARELY USED* Wireless GPS PDAs
      Item # 112371802498652

      Currently US $10.00
      First bid US $1.00
      Quantity 500,000
      # of bids 27
      Time left 7 days, 14 hours +

      Location Washington DC 20233
      Started Apr-02-10 19:24:52 PST

  • That would be a pretty vast amount of paper/trees saved.
    Might actually help re: Greenhouse Gas emissions.

    Nice to see that some parts of the government are doing something (even if it's not deliberate),
    seeing as the Bush-Bot Mark 2 doesn't give a damn.

    (Remember the Kyoto Climate Change Agreement - which everyone else in the world signed?,7369,526607 ,00.html [] alks/ []
    • The idiotic Kyoto Agreement needed so much work that only ONE industrialized country signed it (Romania).
      The fact is that NOT A SINGLE COUNTRY in the EU has signed it, and they are usually even more enviro-radical than some of the moron greens here.

      Get your facts straight.
      • Everyone else said yes they think it should be signed.

        Ratification hasn't happened yet. Romania is an early taker a I suppose.

        EU one step from Kyoto ratification /tech/newsid_1 854000/1854038.stm
        • In other news, a bunch of welfare recipients got together and voted that work should be done, then went back home to watch "Cribs."

          Yes, I get your point.
          Yes, I understand I'm being immature.
    • My error - actually, the US, (& 'everybody else') HAS signed the protocol.

      But the George Bush said that the US will not be ratifying the agreement.

      The protocol will only come into effect if 55 nations ratify it, including the countries responsible for at least 55% of the developed worlds emissions - and since the US accounts for 25% of those emission, it was a fricken big set back.

      But, in Bonn, Germany, in July last year, 178 nations agreed with the protocol, and as of February 2002, 47 of the needed 55 countries have ratified it. otocol.html []
      • From what I read a while ago, it was probably propagandic, the Kyoto protocol would have serious effects on the US economy. Add in that 'developing' nations are exempt, and you have reasons to be concerned. After all, according to the rules, The US would have to reduce total greenhouse gas emmissions to 5% BELOW the total that was emitted in 1990. We'd have to get alot of the show-off trucks (It's a show-off truck if it's never had anything in the bed) and SUVs off the road, as well as seriously clamping down on industry that's grown in the past 12 years. It'd help to get rid of some of the fossil fuel power plants too. Go with nuclear/renewable power sources.
  • I got mine from the factory, refurbished for only 139 bucks... I'm sure in 8 years they'll still have refurbished current ones, and at the rate of depreciation that handhelds have, there should be a huge secondary market.
  • by nurightshu ( 517038 ) <> on Saturday March 16, 2002 @04:10AM (#3172423) Homepage Journal

    Every day, our drivers use their DIAD (Delivery Information Acquisition Device) IIIs to track the millions of packages delivered and picked up. They're wireless, but in the event that they're out of RF range, the information is buffered and then re-transmitted by the DIAD Vehicle Adapter (DVA), which also provides trickle charge capabilities. If all else fails, the DIAD can transfer its batch at the end of the day when it's placed in its cradle at the center.

    We don't currently have GPS in the DIAD III, but the prototype DIAD IVs (which run on PocketPC 2002...ugh) do. They'll not only give drivers who lose their way directions to the next delivery destination, they will also broadcast their location back to the center, which will allow center supervisors and managers to determine more efficient driving routes and coverage areas.

  • The census is only legally authorized to *count* the population. All the other sundry data they collect is the result of a(nother) government power grab. Of course the constitution is becoming more and more just so much bird cage-liner.

  • ... the price will drop by 2009.

    Samsung SPH-i330 []
  • by invckb ( 551932 )
    plus people will commit unnatural acts to sell a half-million computers.

    And what are those unnatural acts, I want to know. Maybe I am in the wrong career.
    • I'll sell Al Gore's left testicle, punch out Prime Minister Chretin (Canadian dude, if you must know...) and sell Mike Tyson to the Arabs as a secret weapon to sell a half million wee boxen. That's plenty unnatural to some folks. Alternatively, I could just walk around town come St. Paddy's in naught but the one-button suit an' the green grease-y paint!! Maybe that option's the better, what do the moderators think?? OK there I feel much better. WHEEEEEeeee!!!
  • That's right, you had hanging Chads last year. Can't you just wait for the inevitable "hotsync reporting errors" and "field information loss" due to battery malfunction? Ahh the federal government. Constantly finding ways to add complexity to any situation.
  • What is more expensive? Paper and pen or PDA?

  • You would be crazy to think we won't be able to have this in 8 years. You can already get cell phones with GPS capability - couple that with a pocket PC and you're most of the way there technologically. Price-wise, well, I guess 8 years should be long enough to beat the price down.
  • sense to re-program them and use them at election time so that nonsense and corruption is avoided.
  • We can easily do this much for this price in such a period of time. Hell, I expect it to be on a regular size wristwatch with skin conductance to a patch into the visual cortex at the speed things are moving. So this story is actually about how unimaginative the Census folks are.
  • Perhaps the U.S. should concentrate more on improving the election system before tackling the job of making the census more complicated!
    • You lose.

      1) The gov't does more than one thing at a time, so your jab at the election system is nothing but a troll.

      2) This is simplifying the system not making it more complicated. The data is all in the computer system to begin with, and there are no billions of papers to be scanned. Less errors, less paper, less time, less complexity.
  • It sounds like they want exactly what I want in one of those handhelds. They sure have come a long way since I worked there many many years ago. Incidentally, The Burueau of the Census was the first agency to put MAPs of nearly every single tiny little down in the country on to computers, including street names, alternate names etc.. So, they have always been on somewhat of a leading edge technology wise. When I was there, they were digitizing the maps onto Tektronix CAD stations that still ran CPM! Personally, I think they will be able to get pretty close to that target price. The Census Bureau doesnt get the respect it deserves..
  • I don't see the $100 price point being met. If they bought today's technology in seven years, they would have a chance ... but today's technology won't be around in seven years.

    What would they need today? A GPS, wireless link and a handheld would run them say $600 (come on, its the government ... we know they won't purchase the cheapest PDA. At a minimum, they'll want color). This would require the current prices to drop by a factor of six.

    Think about the computer industry. The price of a PC has only dropped by about half during the last seven years, and the top of the line models still push the same $4k - $5k price of the original PC. I think that we will see a similar price structure for PDAs in 2010 ... $300 for the technology that they need.

    What holds the prices in check? You constantly need more CPU and memory/storage to perform the same functions!

    The funniest part of this story is that they're trying to limit the cost of the PDAs to $100 dollars (or a $50M total price tag) but they will go spend $200M to have an inept consulting firm write the software ... which then won't work anyway.

    Open source census software anyone?

  • All of the hardware requirements exist today. Looking at Handspring, most list prices now are under $200 - and that's for 2-3 old technology. In 2-3 years, an iPaq 38xx could easily be selling in that range. GPS card today is about $200... Sierra card w/ service could be a wildcard...

    In quanities of 500,000 8 years from now, this can easily be imagined in a wearable device.

    Of course, my one large assumption is that the software will be developed based on an open source OS, avoiding the microsoft tax...

  • by stienman ( 51024 ) <<adavis> <at> <>> on Saturday March 16, 2002 @11:41AM (#3173078) Homepage Journal
    So what they're really saying is:

    We want to give a $50,000,000 dollar contract to a company that can provide 500,000 rugged, easy to use, long lasting (14 hours of active use per charge minimum), PDAs with GPS and wireless communications.

    That should be relatively simple, if they use a free OS. They don't need to be color, though it might help. They do need to be very easily visible. The display can be a larger 1/4 vga screen, which should be much less expensive to make than the current color PDA screens.

    So, let's see... About twice as thick as an IPAQ, and about as wide and long as the old newtons. It wouldn't need to be a real computer, so you could go with only flash to hold the (optional) OS and program code with a compactflash slot for long term storage. The GPS unit will cost $10-15 in quantity, as will the GSM (or APRS, or 802.11b ;-). The LCD, touchscreen and battery will cost $30-40 together, and the processor, mainboard, and all associated electronics will be another $10-20. The enclosure will be maybe $5. Cheap labor would be $1-5.

    Cost each unit:$56 - 80. The R&D (as well as breathing room for unexpected problems) would soak up the last $44-20.

    This could be done in two years, including the development of software that is easily configurable to make census forms and input, enable the communications across the network, etc.

    At the high end, it would leave $10,000,000 for the company doing the development, equivilant to 40 salaried employees for five years at $50,000/yr (yes, some would be more, a few would be less, but the dev time should be less than 3 years, and fewer than 40 employees are needed.) If the company doing the work generalizes the PDA enough(maybe adding local networking to the national networking, etc) then they could sell additional units to other customers. Hobbyists would pay a little for it, but it would mostly stay in the corporate sector.

    That's my bid. I estimate about one year to get the company up and running, one year R&D, two years active development, one year for a limited test run, one year for a production run and distribution, and two years breathing room. The software will allow full remote updating, real time statistics collection, we'll engineer the systems needed to run the entire show, and contract the necessary infrastructure for the wireless data collection.

    • "it would leave $10,000,000 for the company doing the development, equivilant to 40 salaried employees for five years at $50,000/yr
      That's my bid."

      Well, get a project manager and an accountant and refine it, it might work. Here's one tip: double the salary cost to get the actual worker cost (once you factor in HR costs, payroll processing, matching social security, managerial/paperwork overhead, and hiring costs), i.e.a $50k employee costs the company $100k.

      Which goes to show that operating a business isn't something that we comp sci folks are necessarily the best at. But (like you) we can do a good job specing out a project and then let someone modify our numbers!

      (Running a business is always more expensive than it looks. Heck, just _filing_ an IPO is a half-million dollar cost!)

      • There are many other costs associated with running a business - taking care of the employee might be one of the easiest. Also given that this is a government project, there is more red tape to deal with. The thing is, it's possible to do it now at this price, think how much more easily it'll be accomplished with off-the-shelf hardware 3 years from now. Only, they can't wait that long to start...

  • I don't use much paper myself. I think I get ten times as much paper in the mail as I use myself.

    However I sure would miss the paper the few times I use it. Printing out a configuration file for some software or hardware, printing out a chapter of a PDF file. I sometimes need to avoid the noise in my cubicle and noting beats going in to a meeting room, close the door and concentrate of the paper on the table.
    Some times I just unplug my laptop or bring my palmtop, but for some reason it really helps me to see it on paper where I can underline, cross over or write on the paper. If I were to ditch the paper completly the palm or the laptop would need to mimic the paper better. One would need a "palmtop" with the size of a A4 and the resolution and contrast. Then I would need to be able do draw, write on it.
    If you could take the acrobat reader and look and draw on the documents like a real paper it would be great maybe even better. Lets say you had a manual for a program or some hardware as a PDF document where you could "mess it up" by writing and drawing on top of the document. A great feature it would be. You could choose to see it as a clean document or with your own markings.
    Then I might avoid the paper all together.
  • If anyone doubts that in several years we'll have handhelds that have capabilities current devices cannot dream of at prices that you wouldn't imagine, I would point them to an excellent source of information: recent history. (Well, as recent as thirty years ago.) Computers were the size of a building and costed millions of dollars. If someone had told people that in a few decades there would be computers that could fit on your desk and would be hundreds of times as powerful as all of the computers on the planet at the time and be affordable for most Americans to own at least one... you wouldn't have believed him.

    What evidence do we have to suggest that the rate of advancement (which is exponential), will not do the same again? We reached the limit of vacuum tubes, and we discovered the integrated circuit. Why should it be impossible to discover another breakthrough of that same magnitude?
  • What if a handheld breaks? or gets dropped? do we get audited?

    And I'll say "I told you so!" when the wireless points get cracked..
  • Cell phone + built in manditory GPS + Java browser = unit. Someone just has to write the software as an applet for the browser, store some of the data on the phone and have it connect up to the main system every 10 to 15 mins to upload the data.
  • For 50 million we could easily produce this device using readily available components. It's really just an assembly job. You don't have to invent a single thing.

    Start manufacturing the devices in 2008. This gives you 2 years to work out the bugs.

    The software could easily be written today and would run at acceptable speeds on today's PDAs. Keep a small team of 4-5 programers to write the software over the next 8 years.

    The hardware either doubles in speed, halves in size, or halves in price every 18 months. If you wait until 2008 before manufacturing the devices they could be 1/16th the cost. In practice it'd be more like 1/8th given some of the other components like the case and battery. If I can buy a Palm for $200 retail, the cost is well below $100 today. In 2008 the cost of the harware will be below $50 easily.

    Forget wireless - the coverage will never be that great in low population density areas. Instead every night plug the thing into a phone line while it recharges. Upload the collected surveys and download the next day's targets and maps etc.

    The whole device could be assembled from off the shelf components and manufactured in 2008. You'd only need 1 or 2 people to co-ordinate the manufacturing.

    The only problem would be managing the politics. Keeping the specifications of what the device should be capable of to what is needed for the task. Every new person on the government side is going to want their own pet idea implemented.

    Time to start a small company of no more than 10 people. Any takers?

  • Alright, someone has to say it...

    Imagine a beowulf cluster of those babies...

  • I find it interesting that nobody seems to be asking the simple question "Why?"

    The US constitution specifies that US Citizens must be enumerated within every ten year period. Nowhere does the constitution authorize such detailed data collection, nor is it "necessary and proper."

    This update program is called MAF/TIGER []. An optional feature request in this update would involve merging these GPS coordinates with something like the Realsite [] program. This would give a textured 3D model that is geospatially accurate to within 1 meter, based off of aerial, satellite and hand held photography.

    Thinking in paranoid mode I can't help but think of government surveillance via satellite and robot and cruise missiles targeting specific GPS coordinates.

    I'm not seriously worried about anything in particular, but it's the abuse that I cannot foresee that truly worries me. As I cannot conceive of any benefit that this program would bring to the average citizen I have to ask, why are we doing it?

  • The census has driven data processing technology before. In fact, it STARTED it.

    As the population expanded, it was taking longer and longer to compute the results. By about 1890 it was taking almost ten years to complete, and the extrapolation was that the next one would take MORE than ten years.

    So an employee of the Census began designing mechanical sorting and tabulating equipment. He came up with a cardboard card which could be punched with holes representing information, the placed in a "press" where the holes were read electromechanically.

    The first sorting machines involved a human putting each card in the press by hand, causing the lid of the appropriate box to pop open, then throwing the card into the box and closing the lid. (After sorting the cards in each box would be counted.) But with time automatic machines were designed to feed, sort, and count the cards.

    The census put out a contract to have cards made, and the bids that came in were very high. So the inventor went across the street to the Mint and obtained the retired cutting equipment for the previous generation of paper money - which became the dimension of the tabulation cards.

    Eventually the inventor hired on with a business equipment company, designing sorting and tabulating equipment for the Census which found applications elsewhere. A multi-hole encoding for alphabetic information that cards be alphabetized in two passes through a simple machine.

    The inventer was Herman Hollerith, and of course the code was named after him. The company was eventually named International Business Machines, later shortened to IBM. The card was the "tabulation" card, later shortened to "tab" card, but it was commonly known as a "Hollereth card" or "IBM card".
  • half a million hand helds will mean that the tech industry will have a nice (even if it's small) boom.

"The eleventh commandment was `Thou Shalt Compute' or `Thou Shalt Not Compute' -- I forget which." -- Epigrams in Programming, ACM SIGPLAN Sept. 1982