By ,
August 21, 2011 - 7:05 PM
I wish I could say nothing. But with the release of “Windows Phone 7: Mango”, they will forever be connected. This new release of the mobile operating system showcases many performance enhancements and new features. One of which is the ability to snap a picture of any document, and translate the text on the fly to common languages.
This is where all the OCR vendors shout “We’ve made it”, well not really.
I remember the first Mobile OCR application I created. The year was 2004, the operating system was Symbian, and it was a Nokia phone that was just a taste of what was going to come soon in the PDA world. At that time the OCR was actually very actuate. Today’s engines are only marginally more accurate, and only in certain situations. Not only that the engine I was working with was very compact, and able to run directly on the phone (an old ARM processor). Ironically the application I was developing was for the purpose of snapping shots of road signs, OCRing them, and translating them, similar to the goal of Mango’s app. So it would have seemed then that the OCR technology was ready for mainstream mobile use, but it never took off.
I’ve been asking why for many years. Technically OCR was ready, and the phones were not to shabby, so why the delay in OCR on our phones?
Mango of course is not nearly the first modern instance of OCR applications. What makes it unique over all the “Apps” surfacing in the last 1.5 years is that it’s a part of the phone’s operating system, making it even more common place. Their method of deployment is the most popular, capture happens on the phone, but OCR and translation happens in the Cloud. This seems to be ideal as performance of the phone is not jeopardized, and the most robust OCR engines can be used.
This spanning of the technology is exciting yes, but let me tell you the problem. When you ask someone what is cool about this new feature in Mango, they will tell you translation. When you ask them how much they would pay for it, they will say nothing, it’s free.
OCR has been ready for mobile for a LONG time, but the OCR vendors have not. The world of mobile applications is not about core technology, sales cycles, product demos, and sales engineers. It’s about instant gratification, and results driven. Not only that its users are not willing to pay.
To really win in this market traditional capture companies have to change. They have to focus on features. And they need to come downstream and downstream fast. Users don’t care what OCR is. They only care what results they seek. The OCR engine deployed by the Mango product is actually not the latest and greatest, and does not demonstrate the best accuracy. But it’s going to do the job. Far more advanced technology exists, but its producers are un-willing to modernize their approach. It’s time for the capture space to reboot, put their entrepreneur hats on, and stop selling technology and start selling answers.
Rate Post
You need to log in to rate blog posts.
Click here to login.
This post and comment(s) reflect the personal perspectives of community members, and not necessarily those of their employers or of AIIM International
