I attended the Endeca Application Developer training course last month and walked away with a ton of information on this product. Frankly, its quite a bit to absorb in one week. I’ve done a fair amount of work with Lucene in the past and frankly you really cannot compare these two technologies directly. In the world of Endeca, Lucene would just be small but significant part of the larger Endeca solution. Lucene in the Endeca world would simply be the implementation of the MDEX engine and you would need something like Solr to provide some of the other higher search/faceting functionality that Endeca gets you. But overall Endeca is a much larger solution. The pipeline portion alone is quite slick and really could be used as a standalone ETL solution. The pipeline can pull data from a wide array of sources, merge and massage it into Endeca “records” which are then indexed.
Overall I was quite impressed with this product and will look forward to (hopefully) getting to implement a full solution. I am currently working with the Endeca CADK to design a small framework for record adapters.