Re: Barcodes on insect specimens
Date: Sun, 11 Jul 1999 15:32:25 -0400 From: Sprague Ackley <sackley@compuserve.com> Subject: Re: Barcodes on insect specimens Sender: Sprague Ackley <sackley@compuserve.com> To: John Pickering <pick@pick.uga.edu> Pick, Thank you for your thoughtful emails and copying me on your colleagues thoughts. I look forward to meeting you tomorrow, but in the meantime I wanted to put a few misconceptions to rest. 1. By far, the most important aspect of your system is the unique identifier. Make sure it is unique and that everyone uses the same system to insure uniqueness. 2. Code 128 is a widely used and supported linear symbology scannable by anything and printable by everyone. It is in the public domain and a published Automatic Identification Manufacturers (AIM) standard has existed for fifteen years. Being linear, it is substantially larger than any "2D" symbology. If you can use it and maintain uniqueness, great. If not, you should consider using a "2D" symbology. By the way, a Code 128 symbol with 2 alphas and 7 digits is exactly the same length as one with 2 alphas and 8 digits since a single bar code character encodes 2 digits. Similarly, a 10 digit symbol is the same length as a symbol with two alphas and only 4 digits. 3. Code 49 was invented by David Allais of Intermec in 1987. It was immediately placed in the public domain and an AIM standard was developed shortly thereafter. Every scanner manufacturer has at least supported it at one time and could support it at any time if they choose (especially is orders start coming in). Intermec supports it in every terminal and its imager, the 1470. Every printer manufacturer supports it. The world's largest "2D" bar code application in the world, in terms of the number of symbols out there, is for uniquely identifying APS film cartridges with a Code 49 symbol. 4. MicroPDF417 is a new public domain "2D" symbology with a promising future, but without multi-manufacturer scanner support so far. That should change as time go on. MicroPDF417 will store the same information that a Code 49 symbol can store in slightly less space. Very few printer manufacturers support it, yet (none that I am sure of). Typically, symbologies take years for broad based support. 5. No symbology standardized by AIM is "dead" including the very first bar code standards of the early 80's, such as Code 128. Some are less used than others are, but all are supportable and in the public domain. 6. While OCR may at some time be a practical technology, it will require imager based technology, making it as expensive as reading Code 49 or MicroPDF417. However, OCR is prone to error (try telling an F from an E with a scratch through the bottom line). For OCR to be as secure as bar odes, at least two check characters need to be used in every message. You could provide for secure OCR scanning in the future by adding two algorithmically calculated characters to your unique identification number. 7. While it is feasible to have many symbologies in a given application, it sometimes creates difficulties obtaining scanning hardware that supports all of them. For instance, an imager today might scan both Code 49 and Code 128 and sometime in the future, will probably scan MicroPDF417. Some rastering laser scanners today can decode MicroPDF417 and Code 128, and not Code 49, but could scan Code 49 if necessary. Generally, one symbology is easier to support. Again, the unique identifier is the key and the symbology is only the carrier. I hope this is helpful. See you tomorrow. SA
| Discover Life in America | Science | Unique Identifiers & Barcodes | Correspondence | Sprague Ackley - 11 July, 1999 |