PanBI - business intelligence everywhere!
What is PanBI?
PanBI is a collection of analytics modules for existing information systems. For each information system, it provides data extraction, transformation and loading logic coupled with an OLAP schema, delivering OLAP functionality to an unprecedented user base.
It is free software, published under the GNU GPLv3 licence.
PanBI puts Firefox browse history under the microscope
[2010-10-04] As of today, PanBI can be used to analyze Firefox browse history information thanks to the nice people at Mozilla who have started storing the information in a database with Firefox 3.x. Browse history visualization reveals how people use the browser, how and how much of their time they spend in front of the computer, what people look for on the web, how the browser works and to cut a long list short, PanBI is now an excellent tool for people interested in understanding their use of their own time, browser developers, and anyone else curious enough to see how all of it works.
This feature marks the release of PanBI 0.2.1, the first version which can use the wonderful Groovy language to perform ETL tasks. As is customary, a screencast is available to give you a taste of what's possible (problems watching?). Details are available on the PanBI Firefox page...
The official PanBI 0.2 release is out!
After quite a lot of work, PanBI is now available with a host of improvements, most interesting being:
- this is the first publicly available release!
- a subversion repository has been pre-configured by default so that you can validate right away that everything is OK
- reduced technical requirements, simplifying use: basically, users need only have JAVA_HOME pointing to a valid installation of Java
- a really good basis for streamlined release management, which will make future releases much easier
- a rough optimization process reduced the package size by about 50%
- tested on Ubuntu and Windows XP
Lots of great ideas are lined up for implementation in the forthcoming release, but in the meantime, get the fresh 0.2.0 release and enjoy!
New PanBI screencast based on the Iskrica.com dating site data
[2009-08-23] The Iskrica.com on-line dating site has a lot of publicly available data. Because the data is highly structured, interesting, insightful and widely understandable, it has been selected for analysis using the PanBI toolset. As a result, three new screencasts are now available. They showcase how PanBI provides insight into a previously obscure dataset:
- [part 1]: general, simple analysis (15 minutes, 46MB)
- [part 2]: differences and similarities between genders (7 minutes, 20MB)
- [part 3]: age group analysis (15 minutes, 50MB)
The videos are encoded using [OGG Theora]: [VLC] and [MPlayer] play the videos immediately. If you use other players, install the Theora codec and enjoy (Windows users can get it [here], Mac users can use MPlayer, GNU/linux users probably don't need instructions :)).
PanBI gets a Subversion repository analysis module
[2009-05-16] Starting today, PanBI provides analytics capabilities for Subversion repositories. A Subversion repository contains a wealth of information like general and contributor-specific activity trends, the division of work between contributors and much more. In true PanBI fashion, a screencast is available demonstrating initial analytic capabilities on the repository of the worlds No. 1 web server: Apache HTTP Server! Find out more on the PanBI Subversion page...
PanBI debut at the DORS/CLUC 2009 conference!
[2009-05-07] PanBI's first public presentation took place on 2009-05-06 at the DORS/CLUC conference in Zagreb, Croatia. There was an audience of about 50 people who got to see what the project is all about and what it wants to accomplish. The presentation is available here and the demo to which the presentation refers is here.
In the press
PanBI is mentioned in the Bug magazine article covering the DORS/CLUC 2009 conference.
- extracts data from the source database
- sets up a warehouse schema
- loads the data into the warehouse
- defined OLAP cubes
- does all this automatically for a one-click-run experience
- NesterSoft WorkTime
- MantisBT (under development)
The first lines of PanBI code were written in March 2009 and it's first 0.x release is expected before the end of 2009. The current code is available in the mercurial repository, though, so with just a bit of effort, it can be successfully used.
The project is in a bootstrap phase and infrastructural and logistics work is taking place parallel to development.
It can be used with the following systems and applications:
- dotProject project management tool
- WorkTime time tracker
- Subversion repositories
- Firefox browser history
- MantisBT bug tracker, to a degree (still lacks cube definitions and synthetic data generation code).
The OLAP cube definitions are written for Mondrian, which means that they can be used with Pentaho, JasperIntelligence, JRubik, JPivot or other Mondrian compatible application. We currently use JRubik as it's simplest to set up and use and has a very flexible user interface.
PanBI builds on the following platforms and tools:
- Derby database
- Mondrian OLAP server
- JRubik OLAP client
- DDLUtils, SQL Workbench, Schema Spy, ant, mercurial
- GNU/Linux as a development platform and environment
The main place to talk about PanBI are the forums because we want to make all communication which might be interesting to other users or developers public. For very specific and/or private issues, mail will still be used. Other communication media will be considered as the need arises.
This ("other links") navigation is a temporary fix until the hard-working sourceforge crew find the time to correct a problem preventing full editing of the PanBI wiki.
Feature requests and bug reports belong in the PanBI issue tracker.