Searching with Riak Search

It appears that RiakSearch is modeled after Lucene in a variety of ways (https://wiki.basho.com/display/RIAK/Riak+Search):

"At index time, Riak Search tokenizes a document into an inverted index using standard Lucene Analyzers. (For improved performance, the team re-implemented some of these in Erlang to reduce hops between Erlang and Java.)"

"Search queries use the same syntax as Lucene, and support most Lucene operators including term searches, field searches, boolean operators, grouping, lexicographical range queries, and wildcards (at the end of a word only)."

However, there is difference in the scoring function (https://wiki.basho.com/display/RIAK/Riak+Search+-+Querying):

"Documents are scored using roughly the same formulas described here:

http://lucene.apache.org/java/3_0_2/api/core/org/apache/luce...

The key difference is in how Riak Search calculates the Inverse Document Frequency. The equations described on the /Similarity/ page require knowledge of the total number of documents in a collection. Riak Search does not maintain this information for a collection, so instead uses the count of the total number of documents associated with each term in the query."

I am confused by this statement that they don't know "the total number of documents in a collection".

If they were to say: "We don't use the document frequency (# of documents containing this term / total # documents) because we cannot compute (# of documents containing this term) over the entire corpus. Instead we estimate the document frequency using the term frequency (# of occurrences of this term in the corpus / total # terms)." that might be sensible.

But I am not really clear what the current scoring function is.