Innotech's NetResults-CD v.2.3

Emedia Professional, Sept, 1998 by Marla Misek

The problem with Web searching--and indeed, with many quests for knowledge-is that in most cases, what you seek is not necessarily what you find. And while all information seekers would like to believe the Biblical proverb admonishing them simply to "knock" to find the answers they need, the reality is that most search and retrieval engines, whether they be Web- or CD-based, fail to deliver targeted, search-specific responses to what are usually targeted, search-specific inquiries. What's worse, that same information seeker is typically saddled with information that is either frozen in time--having been archived to CD long before the user began his or her search--or so current that the historical or corporate context of the information is lost or distorted.

Innotech Multimedia Corporation is working diligently to change all that. Headquartered in North York, Ontario, Innotech has been designing high-speed search and retrieval tools since 1988 and tweaking its Java-based indexing system since 1996. Its NetResults product suite--which is targeted to Web developers and marketed using a fishing metaphor--features Java-based and 100% Pure Java products promising search and index functionality for Web sites, intranets, and CD-ROM. The comprehensive product family, which is designed to exploit network interoperability, capitalizes on java's "write once, run anywhere" design and has been tested on a variety of platforms, including Windows 95 and NT, Macintosh, Novell Intranetware, Sun Solaris, SGI Irix, and Linux.

Particularly useful to the modern corporate user is NetResults-CD, a multiplatform, multibrowser search tool for Web content published on CD-ROM. With version 2.3--shipping since late May--Innotech has created an applet-free, forms-based approach to information indexing that not only eliminates the security problems inherent to some browsers running applets, but also creates a search environment in which both static, "frozen" information on CD-ROM and current, dynamic information on the Web can be viewed.

THERE'S BAIT, BUT NO SWITCH: WHAT YOU SEE IS WHAT YOU GET, AND WHAT YOU GET IS GOOD

NetResults-CD v.2.3 is relatively easy to install and requires no programming knowledge to operate. Unique to the product is its remarkably small size. NetResults installs no software on the user's hard drive and requires only 7MB of hard disk space for the temporary files used while the program runs. The only catch: your computer must support a version of java Virtual Machine (a.k.a. Java Runtime Environment or Java Development Kit) if you intend to build an index for burning to CD-ROM. Users wishing to search a NetResults-indexed CD-ROM do not need this functionality.

If your operating system is not already Java-enabled, you may download the capability at http://www.javasoft.com. Before testing, I equipped my 300MHz Pentium II with Java for Windows 95, as well as WinZip, a decompression utility from Nico Mak Computing, Inc. Unfortunately, you won't get far with indexing in v.2.3 if your system doesn't support each of these functions.

From there, installation is simply a matter of following the manual, which is both straightforward and thorough. Configured to boot automatically when the CD is inserted, the electronic manual presents a step-by-step tutorial of the product's capabilities, which I printed for future reference. From its initial discussion of where to find a Java Virtual Machine to its final advisory about the complications of burning to CD-R, NetResults' manual is both systematic and comprehensive--in less than 50 pages.

Once NetResults' zipped files are decompressed and installed on your hard drive---a process that takes less than a minute--you're ready to index your content. In this case, I used the content on the Online Inc. Web site (http://www.onlineinc.com), not only because I had instant access to the site's preserved data on Zip disk, but because the company has been experimenting with various search and retrieval engines for indexing its site.

To begin, make sure the content you wish to index resides on your hard drive and that you have opened the proper NetResults Administration interface. Once active, the Administration interface guides you through the process of indexing your Web-based content. The indexing process itself consists of six basic steps: naming your index; editing your index; assigning a path to the location of your content on the hard drive; choosing files to be included in your index; adding "stop" words that will be omitted from the index; and building the index (which activates with a simple click of the mouse). The program also features a number of advanced options for the professional Web developer to incorporate into his or her index.

You'll begin configuring your basic index--which, in testing, took less than ten minutes--using NRIndex Manager. Once created, the Administration interface directs you to define parameters for your index, including the path to be indexed, the files to include, and the words to overlook when searching the index. A "Get Default" feature on the "Stop Words" screen presents a list of common stop words, which can be tailored to your content. After completing these basic steps, NetResults prompts you to build your index using a simple "Go Now" button.