Put Intelligence Where It Helps

 
 
By Peter Coffee  |  Posted 2006-06-26
 
 
 

As observed in the stories of fictional detective Sherlock Holmes, theres more than one kind of intelligence. In an 1893 story by Sir Arthur Conan Doyle, "The Adventure of the Greek Interpreter," readers learn that Holmes admires his older brother Mycroft for having superior powers of observation and deduction—while at the same time saying that Mycroft is "incapable" of applying those powers to the craft of detective work.

"If the art of the detective began and ended in reasoning from an armchair, my brother would be the greatest criminal agent that ever lived. But he has no ambition and no energy," Sherlock tells his colleague Dr. Watson. "He will not even go out of his way to verify his own solutions, and would rather be considered wrong than take the trouble to prove himself right." I found myself thinking of the paradox of an intelligence that lacks the practical faculties of application after speaking last week with Charles Nicholls, founder and CEO of SeeWhy Software.

SeeWhy announces this week the general release of its Enterprise Edition business intelligence product, following an extended period of testing with early adopter sites. As Nicholls told me during our conversation, "We want to change the way that people analyze data. Theres no reason today why people should analyze data thats out of date, but its incredibly difficult to make the data warehouse anything else." SeeWhys approach, he said, begins with a real-time event-driven engine for recognizing and analyzing significant data blips, then couples it into a message-oriented environment that lets developers apply the results close to the point where they can have the greatest leverage.

Nicholls contrasted this approach with what he characterized—and I agree—as a more common business intelligence tactic of gathering data during a process, analyzing it after the fact and alerting someone if that post facto analysis raises any red flags. Instead, Nicholls suggested, an application should be able to invoke the analysis of data in flight against a metrics suite while theres still time to identify likely errors—or to follow up interactively with a user to get the details of whats clearly an exceptional situation.

"If I automate a dumb process, I have an automated dumb process. What I want to do is build business intelligence into the business process," Nicholls said. He compared classical business intelligence software models with the black-box flight recorder extracted from the wreckage of a crashed airplane: What hed rather see is business intelligence technology that functions more like an autopilot that keeps you out of trouble in the first place.

SeeWhy is certainly not the first BI supplier to tell me about the leverage thats waiting to be exploited in real-time data streams. Composite Software, Cognos and Informatica have all earned eWEEK Excellence recognition for their work in bringing analytic capabilities out of the armchair of the enterprise data center to apply them closer to the front lines of operational decision making. I recently spoke with Scott Burk, statistics wizard in residence at closeout retailer Overstock.com, about his use of KXEN Analytic Framework to accomplish that same goal. SeeWhy isnt alone in seeing the "why" to do this—but its a new entrant designed "from the DNA up," as Nicholls put it during our interview, to do it well.

Tell me what you need done well at peter_coffee@ziffdavis.com.

Check out eWEEK.coms for the latest database news, reviews and analysis.

Rocket Fuel