Thoughs on ODTUG Kaleidoscope 2007

I'm currently sitting in the deli cafe at the Hilton Daytona Beach hotel, about to leave and get my connecting flight through to Atlanta. I've been so busy during the past few days that I've not really had a chance to write about the ODTUG Kaleidoscope event that's been running here, but now I've checked out and have half an hour or so to spare, it's a good time to take stock.

First of all, like all these user group events, the main attraction for me is the chance to catch up with friends and colleagues based overseas. This time, I managed to spend a fair bit of time with Dan Vlamis (Vlamis Software), Mike Durran and Mike Donahoe (Oracle Discoverer, Delivers and BI Publisher PMs), Kent Graziano (co-chair of the BI&DW SIG), Matt Topper and Dan Norris (IT Convergence), Wilfred van der Deijl (OraTransplant, winner of the Best Paper award), Lewis Cunningham (of IT Toolbox fame), Lucas Jellema (AMIS) and Jean-Pierre Djicks (OWB PM), they're all good fun to be with and it was interesting discussing latest developments in the Oracle BI industry. The conference was worth it for just this alone.

In terms of product announcements, there wasn't really much new as this year Oracle didn't use the event to launch anything big, as they did last year with OWB 10gR2. One useful session was however Jean-Pierre Djicks look at what's coming up with OWB11gR1 and OWB11gR2; JP outlines the main features on his blog here, and in summary the new features look like being:

• For OWB11g, one new feature will be simplified installation. OWB will now have a central repository installed automatically when the database is installed, and all user-specific repositories will now be subsets of this central repository. This avoids you having to use the SYS password when creating new repositories, and the basic installation will already be done for you as part of the DB installation. The Workflow elements will also be pre-installed and pre-configured.
• OWB11gR1 will also come with functionality to view explain plans for mappings, gather statistics and provide advice on SQL and view tuning.
• Other OWB11gR1 features include support for DML Error Logging, the ability to name an error table in each mapping and have OWB pass error (via DML Error Logging) to the error table if a qualifying data loading error happens - this won't require the Data Quality Option, it'll be free funcitonality. Dimension support will be enhanced, with SCD2 handling fixed, support for hierarchy versioning, and better support for creating materialized views. Interestingly, many of these OWB11gR1 features will be back-ported to OWB10.2.0.3, with a patch available for download on metalink.
• Some information on OWB11gR2 were also made available by JP. Most interestingly, OWB11gR2 is likely to use the JDeveloper/Fusion Development Platform IDE as it's environment, meaning that future versions of OWB will have a consistent Fusion IDE look-and-feel, and will look a lot like SQL Developer and JDev now. All the same functions are there - mappings, project trees and so on, but the framework will be built on the same common JDev frame and it'll be possible for the OWB development team to mix and max JDev, SQL Developer and OWB functionality.
• Other future-looking OWB11gR2 functionality included support for SOA and Web Services, Real-time (CDC) mappings, the ability to import ODI Knowledge Modules into OWB and use them to load and extract from non-Oracle platforms, and support for loading Oracle BI EE repositories in the same way as Discoverer EULs (albeit via a file transfer rather than direct loading of the RPD file

Other than that, in the end I had to dip in and out of various sessions as I've got a ton of work to do in preparing for the next round of BI Masterclasses starting in late July. I did manage to catch talks by Matt Topper (Identity Management), JP and a few others, but for me I had to spend a lot of time in the hotel room on the laptop getting set for events later this summer.

Of course one event I did manage to get to, was the ODTUG BI&DW SIG event hosted by Kent Graziano and myself. One popular topic of discussion was how to keep the ODTUG BI&DW community going over the months between SIG meetings; one idea we had was to try and harness some of the Web2.0 / social networking facilities out there to create a community and faciltate networking between members over the rest of the year. Watch this space for more details, but we're expecting to do something using tools such as LinkedIn or Yahoo! Groups to try and get people in touch with each other.

For me though, I've got to get a cab soon to the airport, I've managed to bag an exit seat on the plane so I'll hopefully be able to build out my examples around BI Server Aggregate Persistence, and if I get time, presentation variables.