In the new BI Expert In Depth report “Information Design Tool vs. Universe Designer — A Primer for Users,” Mike McErlain, CBIP, compares the new Information Design Tool (IDT) with Universe Designer. The new IDT became available with SAP BusinessObjects BI 4.0. Mike noticed that many Universe Designer users are having trouble adapting to IDT after using Universe Designer for so many years and wrote the in-depth comparison to ease the transition. See more of his comments on IDT in the Q&A that follows, and then read an excerpt on IDT’s new file formats. The full report is available to BI Expert subscribers.

Author Q&A 

In your opinion, what is the most significant difference between Universe Designer and the new IDT? I think the most significant difference between Universe Designer and IDT is the Eclipse interface. The Eclipse interfaces opened the door to adding significant new features which enable developers to be much more thorough in their efforts and in some cases eliminated the need for tools outside the IDT to answer questions about the data. What is your favorite feature in IDT? I’m going to say the data display features. It’s so much easier to display data from a column or a table or from several tables across a join structure. Not only can you display the data, the added data analysis functions with distributions for the various values found in the data really help a universe developer understand the underlying data in the universe. The ability to display data across multiple tables to test the joins structures is a very handy feature that should provide confidence in the joins, which should shorten the delivery time on the universe. You mention in the report that a productivity-enhancing feature of the data foundation editor in IDT is the ability to create views of the schema. How do these views help? I think one of the Universe Designer graphical schema palettes is the weakest part of the tool. It’s difficult to navigate when there are more than 10 tables. The developers’ viewing options are very limited and you are constantly changing the magnification of the display to zoom in on a particular set of tables. By using the Views option in the IDT the developer can include any tables necessary in the view to say, highlight a specific data structure. With some of the really large industrial strength universes I’ve seen it will make it much easier for the developer to break it down into much more manageable views that are much easier to understand. What does the new calculated column feature allow you to do? In a nutshell the calculated column can make up for possible oversights in the table design by adding columns to the table within the data foundation. By adding the column to a table you can make it easier to build objects that repeat across different contexts. It is a handy way for developers to add columns to tables that are likely to be used over and over when using data from a specific table. I also find it a handy way to shorten the SQL statement for an object by embedding longer SQL statements as new columns and referencing the new column name in the object’s SQL. What is the business layer and why do you consider it as so important? The business layer is the IDT’s version of the semantic layer which is presented to the report developer/consumer when building or changing a query. I personally believe that the business layer is the most important part of any universe because it not only decides what data in the data warehouse will be exposed to the consumer; it also decides the presentation through the folders structure and ultimately the object’s name, which should be terms consumers easily recognize. Building a solid business layer is an art form that has a significant effect on the universe’s adoption in the user community.

Excerpt From the Report

The following excerpt from the report explains a key difference — the new file formats in IDT: One of the biggest differences between the traditional UNV universes and the new UNX universes is the file format. Universe Designer had a single UNV file that stored the universe components. The IDT on the other hand has four separate file formats to support the UNX universe components. Figure 1 shows the ever popular eFashions universe after the universe was imported. The three file formats .BLX, .CNS, and .DFX are created during the import process of the original .UNV universe from the repository. The fourth file format .CNX is created when a .CNS connection is published to the repository. Here are the new file formats:

• (.cnx/s): Contains the connection information for the UNX universe. The connections once stored in the repository can also be represented by a (.cns) file format that provides a secured shortcut to the actual physical connection. A UNX universe can now source data from multiple connections so more than one .cnx/s files may be assembled into a single UNX universe. More about this feature later.
• (.dfx): Contains the data foundation information for the UNX universe. The data foundation is made up of the tables, derived tables, aliases, joins, contexts, parameters, and lists of values for the universe.
• (.blx): Contains the business layer for the UNX universe. The business layer is the marketing name for the semantic layer, which has always been a big feature of the SAP BusinessObjects platform.
• (.unx): The final assembled file format of the BI 4 universe UNX format is visible only in the repository.