The search results are accurate and are fast. The analytics for what queries are not getting results and what documents get the most hits is nice.
It is (apart from the speed) quite unsophisticated. The entire model is geared to get you to pay more, more, more; which is a bit sad. Bug reports are countered with "we can only fix that on an enterprise server" (which is $1500+ per month).
Easy Implementation and writing/generating Lucene queries are very easy. The implementation is very fast and effective wrappers can be built around it.
I don't like Lucene because of its incomplete implementations, especially for recent state-of-the-art retrieval approaches. When I compare it with the Lemur toolkit, it misses several components.
The search results are accurate and are fast. The analytics for what queries are not getting results and what documents get the most hits is nice.
Easy Implementation and writing/generating Lucene queries are very easy. The implementation is very fast and effective wrappers can be built around it.
It is (apart from the speed) quite unsophisticated. The entire model is geared to get you to pay more, more, more; which is a bit sad. Bug reports are countered with "we can only fix that on an enterprise server" (which is $1500+ per month).
I don't like Lucene because of its incomplete implementations, especially for recent state-of-the-art retrieval approaches. When I compare it with the Lemur toolkit, it misses several components.