Open Source Business Intelligence from Pentaho

Anything that makes it easier for organizations to deploy Business Intelligence capabilities is a good thing. One of the major hurdles, especially for smaller companies, is cost. Tools from entrenched providers like Microstrategy or Business Objects are incredibly feature-rich, but pretty pricey to implement.

Enter Pentaho, a company that sponsors an open source BI platform called (creatively) the Pentaho BI Project. This, unlike other open source projects (there aren't many) provides a top-to-bottom stack, from ETL all the way through presentation. That's a big deal, since it provides the ability to do much more than simple reporting and analysis. (I should note that JasperSoft recently released a competing product, JasperIntelligence, but I'm guessing that it's not as far up the maturity curve as their standard reporting platform).

Some of the more interesting capabilities include automated, role-based bursting, support for multiple output formats (e.g., HTML, PDF, Excel), ad-hoc data exploration (slice/dice, drill-down, etc.), data mining, and support for linking metrics to workflow. At least from a feature standpoint, Pentaho is well positioned against incumbent platform providers.

The efforts of Pentaho and other companies, such as JasperSoft and Actuate (with their open source BIRT reporting platform), are finally providing organizations with credible, perhaps more attractive, options for deploying BI. Removing that barrier to adoption could be just what the BI industry needs to finally push into the "must have capability" space.

Mobile Web 2.0

Okay, I'm setting aside tremendous fear of feeding the hype engine on this one...

Anjit Jaokar (along with Tony Fish) has written a series of articles (here, here, and here) that propose a framework for deploying Web 2.0 capabilities to mobile devices. A lot to read, but thought provoking.

He raises some interesting points, including a method tackling the notoriously painful input/output aspects of most mobile devices (I hear endlessly typing long URLs using a standard phone keypad is one of the key tortures in Hell). To illustrate, he suggests a mash-up of del.icio.us and flickr. I'm not sure I get the example, but I do get his point: Mobile Web 2.0 would add the interesting dimensions of location and proximity to the mix and would, if nothing else, present some interesting opportunities for building Location Based Services (remember that ancient concept?).

Frankly, I'm a bit on the fence on this one. Until there's a breakthrough in useability (and this isn't it), using the web effectively on small devices just won't be possible.