h a l f b a k e r yIt might be better to just get another gerbil.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
The cell phone card reader is embedded in the device with a tiny slit for inserting a business card. once inserted the data is converted into digital information with:
Name
Company
Contact number
Email
date
One can also manually enter a memory string which contains information like where
you met the person, with regards to what etc. Comes in handy for people who end up exchanging business cards all the time.
On line barcode
http://www.barcodemill.com/
Cut and paste possible. [Ling, Oct 04 2004]
Matrix Barcoding
http://www.axtel.com/BarDM.htm
(one example) encode more data than 2d -- may be applicable for cellphone configurations. [dpsyplc, Oct 04 2004]
Card Reader (Baked?)
http://www.amazon.c...ctronics&vi=reviews
An optimized OCR engine quickly and accurately converts scanned business cards into database records and CardIris organizes the information and automatically sorts it into appropriate fields [not_only_but_also, Oct 04 2004]
OCR error rates
http://primerecogni...rime/clean_data.htm
[Ling, Oct 04 2004]
Please log in.
If you're not logged in,
you can see what this page
looks like, but you will
not be able to add anything.
Annotation:
|
| |
some cellphones are smaller than the common business card these days. They also make electronic readers for your computer, and you can also put yur busines card on a cd-r. |
|
| |
A camera phone with OCR could solve this. |
|
| |
Not enough resolution on my camera phone.
How about if every card had a bar code down one side?
It could be either swiped, or a small IR transmitter/receiver fitted to one corner of the phone and the phone swiped down the card.
So it would mean that a standard bar code format might be agreed. Family name, first name, position, number etc.
Mind you, I don't know how long the bar code would be (invents fold out business card). |
|
| |
The software for cellphone integrated-camera barcode recognition already exists (Mitigo CodePoint, Scanbuy ScanZoom, Neomedia and others). 2D codes can directly encode up to 1.8Kb of data. |
|
| |
There are camera phones with OCR as it turns out. The Sharp SH505i is one such phone (this was pointed out me by a good friend whose name is Ling, coincidentally). |
|
| |
Also, link the scanned data with a photograph of that person.
If you have a face for memories like mine, then this would be a great reminder of who you met.
BTW, I've heard that another way of describing card exchanges is "Japanese handshake"
I have used OCR, and it is not fun looking for all the strange interpretations that are caused by specs of dust, and changes in font. Also OCR would need a similar layout of cards, which admittedly could be agreed as easily as the bar code idea. |
|
| |
[Toadinnov], I am overwelmed by the sheer number of codes in your link. But it seems that all the 2D codes need a scanning laser type reader. My first tongue in cheek impression when reading through all the specs was, "Why not go the whole hog and arrange the dots so as to form text?" |
|
| |
No, the Mitigo can do 2D codes and even has a proprietary barcode with higher density than standard 2D codes. If they're still in business, that is. |
|
| |
Yes, you are correct. So in summary, it is possible to have a 2D code on your business card, and read it with an inbuilt camera.
The first side would have the main language,
The second side would have the alternative language,
and the third side would have the 2D code :) |
|
| |
Well there's RFIDs but that'd mean all those nasty servers and databases. |
|
| |
Plus they are about a dollar each |
|
| |
For the moment (actually 37¢ or so) |
|
| |
Just have a standard barcode down the side, but instead of containing all data about you including picture have it be a unique ID at LookUpMyBusinessCard.com (site currently available). The text id could be printed there as well, in case you don't have a reader. AT1844T4 (note the longer-than-needed number, for security) is a lot easier to copy down than someone's name, #, etc. Add a web-enabled cell phone, and you're there. This would have the added benefit of being self-updating. |
|
| |
Of course, it would need to marketing company proof. But as you say, if the code was long enough so that it could not be easily guessed....
I like the idea of automatic updates.
Now all we need is the mobile telephone to be able to accept this code, look up the details, and record internally.
Alternatively, use a PC and update using the program that sometimes comes with the telephone.
However, I wouldn't want to do more steps than are necessary.
How to overcome the problem of area codes? For example, someone may give a card with all telephone numbers prefixed with the country dialling code, but you are in the same country. Some editing required? (Unless the phone is smart and knows which country it is in). |
|
| |
Screw the bar codes. OCR is up to the task. If the problem is your phone res is too low, get a higher res phone cam. (They make 2 megapixel ones now.) |
|
| |
The above comment was printed on a Canon laser printer, at about business card telephone number size, and scanned using an Epson 1670. Finally OCR carried out by Finereader sprint 5.0.
This is the result:
Screw Ibe bar codes. OCR is up to the task. If the problem a your phone res is too tow, get I higher res phone cam. (They nuke 2 megapixel ones now.)
waugsqueke,Marl620M
|
|
| |
Heh. Yep, that's the OCR quality I know and loathe. |
|
| |
Agree with [bristolz] - I think this might be a good RFID startup application with, say, a 3 year gestation. When stick-on low-data-rate ultra-short-range RFID chips are available at sub 1¢ prices, (run a sheet of them encapsulated into business card stock through your laser or offset printer), and micro readers at sub 50¢, then this will be a very fine way to go. |
|
| |
Alternatively use 802.14.4 which is short range low data rate (<200kbps). Or maybe it's time to develop 802.14.5, i.e. really tiny data transfers, really short range, really low power, perhaps driven by piezoelectrical power generation, easily achieved by flexing your business card. |
|
| |
I think there would be security issues with RFID chips. Anyone you walk by could know your name, number, and address. |
|
| |
You'd have to walk VERY close by them, in fact you'd probably have to embrace them firmly. Garden variety passive RFIDs are readable from no further than about 4 inches but much more reliably at 1 to 2 inches. Active RFIDs require a power source, usually a watch battery, and while they are readable at much greater distances they are also one heck of a lot more expensive for both the chip and the reader. |
|
| |
[Worldgineer] - possibly. But if you select the ultra short-range version, which works over just a few millimeters, you should be safe enough, as you'd physically have to touch the phone or reading device up against the card to get the data. Of course, a new breed of technoIDpickpockets could emerge, who just dangle a card-reading wand into your pocket, suck up the data, and scamper off on their merry way. What a fun world we're developing, innit? |
|
| |
But if you go the piezo-power route, the card would remain inacive until actually flexed, when it gets enough juice for a 10-second data transmission before shutting down. |
|
| |
Ok, maybe I'm paranoid of getting telemarketing calls after sitting down at a fast food restaurant. But I don't get how RFID beats a simple printed barcode for this application. |
|
| |
It probably doesn't. Perhaps the best of all would be to have a personal attendant that travels with you and handles introductions and exchanges of information. This other stuff is just so complificated. |
|
| |
[Worldgineer] and [bristolz] are fine in the assumption that it doesn't really make sense to do it this way. But I'd suggest insertion of 'yet' into that assertation. |
|
| |
There are many environments greatly simplified by this technology. For example, ever been to a tradeshow, or, worse yet, had to exhibit at one? A technology like this would eliminate most of the expense and pain of producing expensive machine-readable badges for just a 3-day event, and providing exhibitors with imperfect card-swipe readers. Just slap your own standard business card against the RFID reader, and away you go, to be bombarded with the snailmailed brochure next week. |
|
| |
In the short term, this won't replace the standard drop-your-business-card-in-the-fishbowl-for-a-chance-to-win-a-palm-pilot, and getting the junior marcomms assistant to run them all through a CardScan after the show. |
|
| |
But in 5 years or so, I think this will be baked to a crispiness. |
|
| |
Where they are most significant now is in inventory and mfg velocity contol. |
|
| |
// The above comment was printed on a Canon laser printer ... // |
|
| |
The experiment was done by someone who clearly had an interest in getting a negative result. I stand by my comments. |
|
| |
OCR is clever, but it is not error free. From the link: 14 errors out of 2000 characters (before manual correction). This is using 300dpi on an A4 page (equivalent to 8.6 Mega pixels), under flat bed conditions with very good lighting.
The same link claims 40 errors per page using competitor's software.
Sorry, I don't think OCR is up to it. |
|
| |
I agreee with [Ling] on the OCR quality aspect. |
|
| |