What happened to bar code applications?
"To the curious incident of the dog in the night-time."
"The dog did nothing in the night-time."
"That was the curious incident," remarked Sherlock Holmes.
-- Sir Arthur Conan Doyle
Like Holmes, we're intrigued by an odd silence. With all the noisy debate these days about pen, voice, OCR, and other sexy data input technologies, one of the most important of these technologies--bar coding--never seems to generate much talk.
That silence is hard to explain. Certainly, bar coding has become a virtually universal method for encoding data about things--groceries and toys, express mail packages, library books, magazines and newspapers, retail software SKUs, even the furniture in corporate board rooms. In fact, it's rare these days to find any manufactured item that doesn't carry some form of bar coding.
Moreover, bar coding technology is mature, accessible, cheap, and robust. Under laboratory-like conditions, handwriting and voice recognition are just barely functional; meanwhile, supermarket scanners routinely snatch bar code data off crumpled bags and crooked labels, often through a haze of crumbs and grease. Bar code technology may not be fashionable among software visionaries, but it defines the standards--phenomenally high standards--for accuracy and performance for many of the world's largest data management applications.
Yet, mysteriously, the whole PC industry continues to overlook the bar code market. That lack of interest has left a clear field for vendors of proprietary hardware systems, label printers, and vertical market software products. (To the best of our knowledge, not one general purpose database or accounting package can directly generate bar code output, and there's no system-level support for bar code in windows, the Macintosh, PenPoint, or any other operating environment.)
By ignoring bar code applications, we suspect the PC industry also overlooks some of the best opportunities for productivity gains and cost savings. One small example: The U.S. Postal Service says it will soon offer a discount of up to 3cts on every piece of mail that carries Postnet bar code. (By bar coding much of the mail it already handles, the Post Office itself also has achieved some hefty internal savings.) To us, a quick 10% savings on postage sounds like a pretty compelling story--but so far that story remains largely untold.
We're genuinely perplexed about this lack of enthusiasm for bar code. We've talked to enough developers to be reasonably confident that the technical obstacles are trivial. Research data about the size of the various bar code market segments is weak, but any dummy could figure out that literally billions of objects these days carry bar codes.
The answer, we suspect, is simply that people who develop (and buy) software these days are mostly white-collar paper-pushers. They send memos to each other, schedule meetings, and create various kinds of intangible products. Bar code isn't especially useful in this paper world; thus, it's easily overlooked. Or, worse, it's dismissed as a "vertical market application"--that is, boring, difficult, and unprofitable.
But there's also a growing consensus in the software industry that traditional white-collar markets have reached a point of near-saturation. To sustain growth, developers almost certainly will have to create products for millions of new users--retail store employees, assembly workers, mailroom clerks, route drivers, parts handlers, and the like. Bar code may not be the only technology that opens the door to these markets, but we suspect it's a great place to start.
|Printer friendly Cite/link Email Feedback|
|Date:||Dec 26, 1991|
|Previous Article:||How to test usability.|
|Next Article:||Is Apple irrelevant?|