Content Copyright © 2006 Bloor. All Rights Reserved.
I just recently wrote that Mobius has the best solution I have seen for spreadsheet management. And, wouldn’t you know it, just a week or two later, I run into another, rather different but equally valid, way of doing much the same thing. Worse, I already knew the product supported Excel (and StarOffice for that matter) but I hadn’t put two and two together.
The product I refer to is Informatica PowerCenter and, specifically, its unstructured option. Now, I have written about that previously and I was well aware that it supported the ability to read Excel data though I am not sure that I knew that it can also write (or write-back) to Excel spreadsheets. In other words, you could use the data lineage, impact analysis and similar facilities of PowerCenter to track changes to your spreadsheets and you could use the product’s transformation capabilities to consolidate, merge and otherwise manipulate spreadsheet information.
Actually, this isn’t quite as broad a capability as that of Mobius since you can still play around with the spreadsheets outside of the PowerCenter environment and, even within it, you wouldn’t have features such as cell-level locking, but it is powerful nevertheless.
In practice, this unstructured option competes with DataStage TX in terms of semi-structured data but it also does a lot more than IBM’s solution in terms of its support for unstructured data. Bearing in mind also that PowerCenter is in the process of being integrated with Composite Software’s EII (enterprise information integration) solution then it also potentially competes with WebSphere Information Integrator. However, it is some way off being a fully-fledged competitor to the latter. For example, it doesn’t have any search capabilities and it does not have native access to content repositories—you can access them but there is no comparable facility to, say, the VeniceBridge product that IBM acquired and merged into Information Integrator.
Of course, you can see the point towards which both companies are converging: a single data integration platform that provides all the sorts of facilities that have been mentioned (plus, of course, data quality). However, I don’t think it ends there. While it is a topic for another day, I also think that it makes sense for data integration platforms to include the facilities you need for non-persistent MDM (master data management). That is, where either you just want to build a view across the relevant data or where you simply want to store pointers that reference source systems, but in either case you do not want a master copy of the data. If you go beyond this you start to get very close to being in the applications space and then you are competing with the likes of SAP, which Informatica at least is clear that it does not want to do.
Going a step further, once you have put all of the above in place then you basically have a data services bus, information services bus, call it what you like. IBM is initially targeting this market with its Information Server while Informatica will be building out from PowerCenter. Notice the different stances—integrated point solutions from IBM, a more monolithic approach from Informatica (the company’s basic method is that everything should, ultimately, be integrated at the engine level not just the metadata level)—but that too is a discussion for another day.