In the early years of personal computing, there were standalone word processing, spreadsheet, presentation and drawing applications. These were quite useful, but they weren't too integrated. If you needed to put together a business document containing text, tables, graphs and illustrations, you had to create and print each component separately and then manually combine them. It was easy to get the components out of order, and page numbering for large documents was often a tedious, nontrivial task. The amount of redundant work effort was huge.

In response to this situation, several vendors integrated their business applications into software suites. Today, suites such as Microsoft Office provide extraordinary capabilities to the average user. Now, we can create documents that contain a variety of objects, all within a single application. I can write my column in Word, go to the insert menu and bring in an Excel spreadsheet and PowerPoint presentation without ever leaving the Word document.

The end-user data access/analysis software industry is starting to reap similar benefits from integration. We've seen tremendous advances in the end-user access technologies in recent years – beautiful visualization capabilities, sophisticated exploration and mining techniques and, of course, the ever more elegant online analytical processing (OLAP) tools. There is a new, interesting and exciting direction for these technologies: third-party tool integration – that is, the ability to access different vendors' tools through a single seamless and consistent interface.

The Need for Third-Party Tool Integration

Last year, DM Review (July/August 1999) printed a series of articles dealing with five communities or types of users of the Corporate Information Factory – farmers, explorer, miners, operators and tourists. Each community has its own set of functionalities and technologies, i.e., OLAP for farmers, exploration warehouses for explorers, operational data store for operators, etc. This model is useful for organizing the users into understandable categories, but it can mislead the reader into believing that the five types are mutually exclusive when, in fact, they are not. Most of our community members don't stay in one community or perform a single role all of the time. They tend to change roles as requirements demand and to move around from community to community.

For example, explorers may discover an interesting finding and begin producing a standard report on that finding every week (just became a farmer), or the explorer may try to determine if his finding is statistically significant (just became a miner). The operator may want to see how productive his unit has been over the past year (just became a farmer) and so on. The ability to easily switch gears and perform a different form of analysis is driving the need to have the same form of integration between different tools that we've seen in Microsoft Office. In this case, third parties are agreeing to package their tools together and offer them as a unified set. The key to this integration is a set of standards to which the vendors can adhere.

Third-party tool integration is much more than just agreeing to a standard data interface such as OLE DB or ODBC. It is the actual incorporation of a tool's functionality inside of another tool. There are several examples of vendors having this new form of integration, but I will illustrate where the industry is going using the following example.

This example shows the integration between six different companies' products. These companies are:

  • Knosys (www.knosysinc.com)
  • Maximal Innovative Intelligence (www.maxsw.com)
  • MapInfo (www.mappinganalytics.com)
  • ParkerSoft (www.parkersoft.com)
  • DWSoft (www.dwsoft.com)
  • John Galt (www.johngalt.com).

Let's assume that you want to create an interactive annual report for your corporation, one that contains not just an analysis of the prior year's activities (sales of products, best customers, top-performing sales representatives, etc.) but also shows graphically what market segments were hot, what trends were spotted in the marketplace and, finally, includes sophisticated forecasts for the next year.
The starting point is Knosys ProClarity Analytical Platform. This is a Microsoft-specific ActiveX/COM environment that uses its own object orchestration system to permit extensibility – that is, the ability to insert additional COM objects into the system without any prior knowledge by Knosys. Having an architecture similar to this means that the analytical application developer can quickly incorporate other vendor technologies or develop new objects that work with the other components.

We begin the creation of our annual report with the OLAP capabilities of the Knosys component plug-in, ProClarity. We create a standard series of analytical queries of sales information, permit the drill through to lower levels of detail and then create the briefing book, stored in XML for Web viewing. So far so good.

Now, by simply pulling down a menu bar from within the Knosys environment (similar to the Office suite "import" capability), we bring in a new functionality such as the geographic capabilities of MapInfo's MapX tool to show sales or customer information by sales region, state, country, etc.

Next, we select the Maximal Innovative Intelligence product, Max, to give us guided animated sessions to spot hidden trends and patterns. We can view multiple aspects via a graphical interface of the year's activities and quickly find answers to questions we may not know about.

Finally, we develop the forecast for the next year by invoking a tool such as ParkerSoft's ezForecaster or John Galt's ForecastX. We can create a simple time-series forecast or do more complicated forms of forecasts such as outlier detection, seasonal pattern recognition or causal forecasting and add it to our annual report.

We have now created a very complex annual report with relative ease, but one last question remains. To put the finishing touches on our report, we must ensure that the data our report is based on is of the highest quality. We need the ability to audit the numbers being delivered from the data warehouse to the data mart. To do this, we simply select our final tool from the Knosys menu bar – DWGuide Navigator from DWSoft. We now extend our functionality to include specific audit areas allowing us to review when and by whom the numbers were aggregated and to view the source tables from which the data was pulled.

Now we are finished with our very complete, very interactive annual report. The report was created as easily as we create our complex Word documents. Our productivity is greatly enhanced; and, best of all, we only needed to invoke a single tool to get the benefits from all six tools.

More and more tool vendors are going down this road by incorporating analytical capabilities from third-party vendors into their own. Plug-ins, unique analytical capabilities and extended base functionalities allow organizations to solve a variety of problems with ease. This significantly reduces the complexity and support costs for the end users as they "roam" through the different villages of their business intelligence communities.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access