Intelligent Systems Report

January 1998, Vol. 15, No. 1

1997: The year in review



It finally happened. After years indeed, decades of waiting, the legendary "killer app" for the intelligent systems industry made its long-awaited debut in the mass-market arena. Everybody by now has at least seen one, and anybody with a child probably has one in their homes. Yes, the "killer app" we're speaking of is the Tamagotchi, also known as the Giga-Pet, an honest-to-goodness application of artificial life technology embedded in a consumer-oriented gadget.

Ironic, isn't it, that more than 40 years removed from the Dartmouth Conference that officially launched artificial intelligence as a movement, that it would be an electronic version of the pet rock that would ultimately capture the largest share of the public's consciousness?

While the Giga-Pets will quickly fade from the scene as all such fads do, 1997 saw another fascinating application rooted in the gaming world grab its share of headlines: The victory of IBM's Deep Blue chess-playing program over Grand Master Garry Kasparov. True, Deep Blue does not actually have an AI component, but in the public's eye, if a computer can beat the best chess player in the world, it must have some kind of intelligence.

Mostly, Deep Blue did what it did by virtue of a supercomputer capable of calculating 200 million chess moves per second, as well as a program specifically designed to foil Kasparov. The IBM team ended up claiming a $100,000 prize for being the first computer chess program to defeat a reigning world champion; Kasparov, meanwhile, appeased his wounded pride by selling off a virtual reality company he helped found ParaGraph International to Silicon Graphics. If you can't beat 'em, join 'em.


The rich get richer
Two of the largest software companies in the world Microsoft and Computer Associates dove headfirst into the intelligent systems arena in 1997, endorsing a number of technologies with their corporate checkbooks. Microsoft, for instance, invested $60 million in various speech recognition companies within the space of a couple months, the bulk of that money going to Lernout & Hauspie ($45M) and Wildfire ($12M). Microsoft's goal is no less than to make its Windows platform thoroughly speech-enabled.

Virtual reality is also of strategic interest to Microsoft, as the company acquired Dimension X, a provider of Java-based platforms and tools for VR content creation.

Computer Associates, on the other hand, was pretty quiet for most of 1997, right up until it introduced its Jasmine object-based database platform. After that, CA quickly made up for lost time by acquiring one company outright (AI Ware) and signing technology deals with two others (Superscape and Excalibur Technologies). Within a few weeks' time, CA gained access to neural network, VR and knowledge retrieval technologies developed by some of the best-known players in their respective fields.


An epidemic of merger mania strikes
Microsoft and CA were by no means the only software companies looking to pad their annual reports with acquisitions of intelligent systems vendors. A veritable merger mania swept through the entire advanced computing arena in 1997, and an incredible number of familiar names disappeared from the radar screens via takeovers.

Consider just this roster of AI pioneers, all of whom were acquired in the past year: Exsys, Information Builders, Kurzweil, Netbot, NeuralWare, ParaGraph and Virtuality, as well as the above-named AI Ware and Dimension X. A couple of these companies vanished without a trace inside the corporate walls of their new parents, while others remain as divisions or subsidiaries. All, however, have lost their independence, and it remains to be seen whether the industry stands to gain or lose by all of these acquisitions.

Just as fascinating to watch have been the efforts of some of the remaining AI companies to grow larger themselves via acquisitions. Excalibur, HNC, ILOG, Neuron Data, Verity and other AI firms added new technologies and markets to their rosters when the fit seemed right, or as in the case of L&H's takeover of Kurzweil, when it was possible to eliminate a major rival. As long as Wall Street continues on its incredible hot-streak, expect to see more acquisitions in the AI industry. Simply put, right now is a very good time to be an advanced technology company (see sidebar, page 3, for a scorecard of all the corporate maneuverings in 1997).


Look into the camera, please
A technological trend that emerged from the R&D backwaters and found some commercial success in 1997 was biometrics, a pattern recognition-based application for personal security identification. Two companies that we followed closelylast year are Miros and IriScan.

Miros' TrueFace technology verifies users' identities by recognizing their face. The company's CyberWatch product uses intelligent software and a small video camera mounted on a PC monitor to verify computer users when they try to access protected data on their computers. CyberWatch then determines whether the user's face is the same as the face image stored in its memory. If not, access to secure data is denied. Mr. Payroll, a supplier of 24-hour automated self-service check-cashing machines, is now using TrueFace technology to verify its customers.

IriScan holds patents on iris recognition, which identifies individuals by computer analysis of the patterns found in the eye's iris. While the technology has obvious applications such as those mentioned for CyberWatch (cash machine transactions, computer and Internet access, credit card transactions), the true value of biometrics may lie in areas somewhat off the beaten track. For instance, a racing association is using iris recognition to identify and keep track of thoroughbred horses, to thwart the high incidence of thefts and horse-switching.


The reality of virtual manufacturing
Now that the hypestorm over virtual reality has died down, the technology behind it has moved to the forefront and is generating a whole new wave of interest in a virtual manufacturing trend known variously as concurrent engineering or collaborative computing. Basically, the idea is to enable different groups to create individual parts of a mechanical design at different geographic locations, and at different stages of the design process.

Adding VR to the mix enables manufacturers to create, visualize, interact with and analyze these designs, utilizing 3-D CAD data. The goal is to improve overall quality of design as well as time-to-market, while reducing the expenses associated with design changes made late in the game.

So far, it has been the extremely well-heeled manufacturers automakers, shipbuilders, nuclear power plants as well as various federal agencies, such as NASA and NIST, that have invested their time and talents in virtual manufacturing. While the cost of hardware and software continues to drop, these concurrent engineering initiatives are by no means small-scale affairs, involving as they do enterprise-wide efforts. However, several recently introduced products are aimed at the mid-sized market, so expect to see a substantial push by technology vendors emphasizing the return-on-investment afforded by virtual manufacturing.

And with that, we wish you all a very happy and successful 1998!


Web Site � Copyright 2020 by Lionheart Publishing, Inc.
All rights reserved.



Lionheart Publishing, Inc.
2555 Cumberland Parkway, Suite 299, Atlanta, GA 30339 USA
Phone: +44 23 8110 3411 |
E-mail:
Web: www.lionheartpub.com


Web Design by Premier Web Designs
E-mail: [email protected]