hi, Though the code has supported for customizing scoring on a per-field basis in using <similarity/> in a schema's fieldType and we can configure our custom similarity factory in schema, we can't configure the default similarty and it is hardcode in SchemaSimilarityFactory.
If we want to use another similarity as default similarty instead of DefaultSimilarity provided by lucene, we must to write another similarity factory to do this. Therefore, i think we can make default similarty configurable. Any comments is welcomed. HeXin
