It’s a big day for those of us who work with the Oracle Endeca Information Discovery (OEID) Platform closely. Doubly so, for folks who love to work with unstructured data.
As of last night, the newest release of the OEID Platform 2.4 is available on Oracle’s E-Delivery site (edelivery.oracle.com). Last night and this morning we’ve been taking it for a spin and what follows is a quick summary of “what’s new?” and a couple things to watch out for.
There will be a formal release announcement coming from Oracle shortly but, from what we’ve gathered, these are some high-level highlights of what is included in 2.4. This is not meant to be comprehensive, just what we’ve come across so far:
- New Text Enrichment support for German, Spanish, French and Portuguese
- Upgraded Version of Text Enrichment from 5.0 to 5.1
- A built-in Integrator Wizard for loading data directly from Oracle BI Server
- Custom Text Enrichment Pack for Social Media – this looks awesome!
- New JSONReader Component – Perfect for integrating Data as a Service content
- Comprehensive Service Pack that encompasses all fixes made to OEID Studio 2.3
That’s just what we’ve come across so far in the last 24 hours, so I have to imagine there’s more that we haven’t gotten to yet. Pretty awesome.
Upgrading OEID 2.3 to 2.4
As previously mentioned, we’ve already sat down and spent some time with the product. The installation was mostly smooth sailing but we did run into a couple small hiccups along the way:
First, from Googling around, this error could be environmental but I ran into an issue attempting to run the Standalone Integrator installation, which I traditionally prefer to the QuickStart:
After trying a number of solutions such as uninstalling older versions (protip: it appears you can run 2.4 alongside 2.3 which is huge for consultants), I finally punted and opted for the QuickStart installation. For whatever reason, the QuickStart install played nicely with my machine where the Integrator installation would not. Your mileage may vary.
Second, we have a ton of sample projects, proofs of concept and other interesting Endeca instances at Branchbird. I immediately decided to open one up and bootstrap up to 2.4. The upgrade process is well documented in the Migration Guide and pretty typical for those who have been through an OEID upgrade before. Again, this process was mostly smooth and everything worked great in my Integrator graph until the execution of Text Enrichment. My existing Text Enrichment Manipulator failed, throwing the following error:
java.lang.NoSuchMethodError: <init> at com.lexalytics.salience.Salience.initializeIds
<snipped for brevity>
Having worked with Lexalytics in the “raw form”, meaning directly to the Salience APIs without a wrapper or a component, I figured this had something to do with the version that I have installed. It turns out that upgrading to the new version of Text Enrichment is actually required as part of the 2.4 installation if you employ the Text Enrichment features in Integrator. As an aside, version 5.1 is way more stable than version 5.0 and we’ve been “backporting it in” to some of our existing customers to improve performance. If you’re looking at Social Media data, you’ll also want to make sure you’re leveraging the new Social Media data collections for your enrichment.
After tackling the Text Enrichment issue, my graph loaded with no problems whatsoever and, as a result, we are up and running on 2.4 over at Branchbird HQ. We’ll certainly have more to say on OEID 2.4 in the coming weeks but, so far, we’re really impressed and it looks like a huge win ahead of OEID 3.0 coming later this year.
Feel free to leave anything else you find interesting in the comments below or drop us a line at firstname.lastname@example.org.