Ian Kallen wrote:
Elias Torres wrote:
A few presentation notes that I think are applicable to our work:
Tags don't map well to SQL. Sometimes you can prune based on usage -
only index the first few pages for example. This keeps indexes small and
fast."""
My suggestion: use the minimum SQL required to assign synthetic IDs and
map relationships. But for querying a large corpus, don't bother with
SQL. The lucene jar is already in roller... use it.
I am not particularly well versed in how lucene works or in the details
of indexing, but that sounds like a *really* cool idea. definitely
something we should look into further.
-- Allen
