The Next Big Thing
Posts about next generation technologies and their effect on business.

The Cyborg Advantage in Business

This month's Wired magazine has a story titled: Advantage: Cyborg, that talks about why an average human with a very strong computer based smarts is the best combination in chess. The people can blend the number crunching of the compute with their innovation and creativity - knowing how and when to rely on machine intelligence.

One of the areas I've been talking about for a while is the intersection of the abundance of data (that people just can't handle) with the abundance of computing capabilities (available from cloud techniques) to define a new plateau of business value generation. This blending of the computer's capabilities and the individual's expertise is what I mean by allowing people to focus on turning anomalies into opportunities and allow the automation of "normal". Focusing the expertise where it is needed, instead of being wasted on normal activities.

This is where pattern recognition and human intuition can give an organization a significant advantage.

StreamInsight – pulling CEP upstream?

Last month Microsoft StreamInsight went public. This software provides a platform for developing and deploying complex event processing applications based on recognizing patterns and enabling an organization to act upon them. "Its high-throughput stream processing architecture and the familiar .NET-based development platform enable developers to quickly implement robust and highly efficient event processing applications. Typical event stream sources include data from manufacturing applications, financial trading applications, Web analytics, or operational analytics. StreamInsight enables you to develop CEP applications that derive immediate business value from this raw data by lowering the cost to extract, analyze, and correlate the data and by allowing you to monitor, manage, and mine the data for conditions, opportunities, and defects almost instantly."

Complex event processing (CEP) is something I used to interact with the Tandem folks (ZLE) on back in the mid 90s. In the mid part of this decade, it generated a lot of hype and then fell into the hype cycle Trough of Disillusionment. Hopefully tools like StreamInsight will help drive in back up the slope of enlightenment, using the significantly more powerful computer resources available today. I fully expect this kind of pattern recognition to be built into most cloud services. 

Three laws of robots actually a human restriction

In Computer magazine, there was a recent article Beyond Asimov: The Three Laws of Responsible Robotics. This article addressed a perspective I've always had - Robots will need to have significantly more "perceptual and reasoning capabilities" in order to even approach being able to follow the three laws.robot

The three laws are more constraints on how the robot's designers should think about their creation than they are something that constrains the device itself. The authors recognize that fact.

This article proposes "an alternative, parallel set of laws based on what humans and robots can realistically accomplish in the foreseeable future as joint cognitive systems, and their mutual accountability for their actions from the perspectives of human-centered design and human-robot interaction."

I found it well worth reading since many organizations are focused on automation and we've mentioned many times in this blog the intersection of computing expansion (i.e., cloud), interconnection (i.e., Internet), and significantly more data (i.e., sensors). These are the foundational needs to have greater robotic capabilities and improved business value delivery.


The issues with robot interfaces being too human

There is an article in this month's Smithsonian magazine about robotics and the issues with what they can do, and the problems people have when robots look too human. This is a good article for those interested in robotics or in the numerous robotics projects currently taking place.

The article also talks about the "uncanny valley," where a machine looks just human enough to be disturbing. This has been studied quite thoroughly and seems to be independent of culture and age.

Even in this mundane example of a robot that focuses on helping people loose weight, the reviewer thought it was a bit "creepy".

Business seems to not have much of a problem with this in our "robotic interactions" since most of those are limited to using IVR systems (and we know about their limits.).  As of right now, we are developing more sophisticated kiosks and more powerful telepresence techniques.

With the large processing engines capable of understanding speech (like Microsoft does with the mobile implementation of live search), it means greater access to data and a wider range of contextual understanding. This is one of the areas that should advance rapidly after the economic downturn. It is also an area where the interaction could creep into the uncanny valley, even if there is no physical presence.

Agentry and birds

Anyone who has been working with IT for a while recognizes that there are pendulum-like shifts from centralized to distributed computing. The centralized approach is more orderly with greater hierarchical control. The distributed approach tries to be flexible and more efficient with time - at the cost of more hardware. This parallels the post earlier about why certain fads follow other IT fads.

When I think about centralized vs. decentralized, it is usually based upon where the data resides. In a mainframe or most Web solutions, the data is under central control. Under client server solutions, many times the data resides in different tiers with specialized processing on each tier.

As we move into an age of autonomous agents and service oriented architectures, the degree of information and logic distribution becomes very distributed.

Computing will be more of a group activity that could be compared to a flock of birds. Birds generally have a common goal, but no clear leader. They can work together using their own set of rules, but within a common framework. Each software agent or service will provide its capability to the enterprise but have its own set of logic.

This somewhat chaotic system of systems is one of the reasons EDS' Fellows are constantly discussing the role of simulation in this future world. It may be the only method of understanding the behavior at an enterprise level.

Follow Us
About the Author(s)
  • Steve Simske is an HP Fellow and Director in the Printing and Content Delivery Lab in Hewlett-Packard Labs, and is the Director and Chief Technologist for the HP Labs Security Printing and Imaging program.