|
|
|
|
Cut Through the Clutter of Enterprise Search
by P.G. Daly
You'd think by now in the age of Web 2.0 that searching for information within the enterprise would be a slam-dunk. Users would easily and effortlessly find the most accurate, up-to-date information they needed. Unfortunately, for a variety of reasons, that is still not the case (for the most part).
There's the old question: "If a tree falls in a forest and no one is around, does it make a sound?" When it comes to enterprise search I have a new question: "If no one can find the information they need when they need it, does it even matter if the information really exists?"
In the Internet world, search is king. Organic search results get better and better as more Web sites employ search engine optimization (SEO) techniques. Users get savvier about how to formulate their search criteria. Pay-per-click advertising delivers instant results every time. The big search engines (with Google leading the pack) employ super-secret, very complex algorithms to determine what results are displayed when someone searches. So, why does searching within the enterprise fall short?
If your experience is much like mine, you know the frustrations of finding what you need when you need it within the confines of the wild world of a large corporate Intranet. With all the portals, Web applications, file systems, content management systems, and other fancy repositories you would think that the information you are looking for just has to be there somewhere. Problem is you can't find it. Or, you find outdated, irrelevant information.
While the technologies are the same -- a combination of systems, Web sites, applications, databases, and files -- the way they are used, stored, and ranked are different. On the wild Web, search technology can employ criteria like popularity, link exchanges, Web traffic, social networking, and tagging to refine search results. Unfortunately, much of the same criteria and metadata that drives the Web is nonexistent, irrelevant, or inconsistently used inside the enterprise.
Two Paths to InformationJust as on the Internet, you have two types of potential search within the enterprise: browsing and searching. Browsing is akin to the Yahoo type directory where information is filed into browsable cabinets that people can navigate to find what they want. Directories can be built by a combination of manual review and filing (think: someone files it in the "right" cabinet) and through a systematic use of metadata and taxonomy on all documents.
Searching is what we've all come to know as "googling." You type in a word or phrase and a bunch of algorithms chug in the background and spit out the most relevant results (as defined by a "secret sauce" algorithm). Both approaches have their value depending on what a user's needs are.
Where things break down within the enterprise is in the sheer volume of data in the corporate network and the wide variety of ways the information is stored. Unlike the Web, which is primarily driven by Web sites, links, and some cross-platform files (such as PDF), the enterprise is a free range of every file type you can dream up. So, any search tool needs to account for thousands of possible ways the information might be stored with key information (text and content) often hidden within the bowels of a document embedded in a file system or database somewhere.
To make matters worse, the motivation for classifying, linking, and tagging information often vanishes within the enterprise. Even if you attempt to create a standard metadata, tagging, and taxonomy procedure, the vast amounts of information makes it prohibitive to do thoroughly. If you're incredibly lucky, you tag the most important information and hope for the best.
Tools to Help You SearchNeed some help with this game of information hide and seek? Enter the wide range of search tools available to use within businesses. Some are provided by large, well-known vendors while others are offered by small, relative unknowns. You can be guaranteed that because of the inherent challenges in this search game that there is no one-size-fits-all solution. In fact, many of the search solutions are highly niched for narrowly defined purposes. You may find yourself using multiple solutions within your company to serve different purposes.
Let's take a high-level look at some of the major players in enterprise search. (Three of the following search tools were nominated by readers in Intranet Journal's Product of Year awards, voting for which is going on now.)
GoogleGoogle is synonymous with searching the Internet. After all, "Google" has even become a verb. Google also offers an enterprise search solution called Google Search Enterprise. Google's search engine crawls your Intranet and other content, creating a master index that is at the ready to retrieve results when someone searches.
Information on how the product works behind the scenes is as closely guarded as the "secret sauce" in Google itself. With support for more than 220 file formats it marches in step with other vendors in terms of what types of files it supports. Google Search Enterprise provides the end user with results clustering and source biasing so that you can expect relevant results, and those from authoritative sources (as defined by your administrator) are more likely to appear in their results.
The Google product Web site claims that in order to find the highest quality and most relevant documents, Google factors in more than 100 variables for each query. So, once again, the "secret sauce" is at play even though we don't know what those proprietary 100 variables are.
ISYSISYS:web 8 is the company's latest suite for enterprise search, navigation, and discovery. Whether used across networks using the "desktop" version or via your Intranet using the "Web" version, the same core technology is at play.
ISYS boasts support for 200 file formats and 60 languages and multiple query methods. So, you get the advantage of browsing taxonomy while also being able to search on the fly. It offers on-the-fly categorization to make use of metadata inherent in the file and its storage location while allowing administrators to further categorize information using additional metadata. Search results also allow users to navigate between and within search results and documents to take them right to the information that matches the users search terms.
AutonomyAutonomy is a leader in Meaning Based Computing. On the company's Web site they define Meaning Based Computing as enabling computers to understand the relationships that exist between disparate pieces of information and perform sophisticated analysis operations with real business value, automatically and in real-time. That's a mouthful for saying that they develop products that help you find and relate the vast amount of unstructured information that exists within your company.
Autonomy IDOL is the foundation of all the Autonomy solutions. It is a platform for understanding the meaning and significance of disparate information. A variety of modules are included in the server to help you handle taxonomy, linking, categorizing, and retrieving information. The functionality of the server is realized through its applications which include desktop search, CRM, portal, Web interface, and collaborative networking.
Fast Solutions focus on the information needs of companies to help them answer the "who, what, when, where, and why" questions to support business decisions, compliance, and investigative activities.
FAST Enterprise Search Platform (ESP) is the foundation of the company's products and is designed to search large networks for numerous file types. You can extend the ESP platform with other FAST offerings to provide the same search capabilities to your users regardless of whether they are mobile or connected to their desktop providing a single access point to all your corporate information. Specifically designed for the needs of search within large internal organizations, they offer their FAST Intranet Solution.
IBMIBM has many fine business tools and enterprise search is no exception. IBM OmniFind is IBM's solution to solving your search woes.
There are 4 different OmniFind products to choose from. Enterprise Edition, Enterprise Starter Edition, Discovery Edition, and Yahoo Edition. The Enterprise (and Starter) editions are optimized for Lotus Domino and WebSphere portal environments. As you might expect, the Yahoo Edition is a collaboration with Yahoo and offers users the familiar search and browsing interface of Yahoo for your company's Intranet, local, and remote file system content. Formerly known as WebSphere Content Discovery Server. The Discovery Edition utilizes IBM's Content Management and Discovery capabilities to provide contextual understanding capabilities to content search.
The market and technologies are still in flux when it comes to enterprise search, and I would expect that to continue for some time and volumes could be written about the topic. A quick visit to Enterprise Search Center yields more information than you probably need.
If you've been hoping for a magic bullet that allows users to use one perfect search tool to find any information they need quickly and reliably across the enterprise, you're just going to have to keep waiting (and hoping the fantasy will come true, much like the truly paperless office). The good news is that a variety of tools already exist to help you get your arms around your search needs and will get you headed in the right direction.
|
Document Saved Successfully
|
|
|
|
|
|
|
