This week, our survey asked participants to indicate which was the primary repository of content for their search application. Most often, respondents cited databases and the public internet as the data sources they built their apps for. Here are the responses.


Interestingly, none of the respondents indicated they were building apps which had as their primary source of docs ‘crawling my company’s ‘intranet’ for HTML docs’ or using a ‘NoSQL-style data store (CloudDB, Memcached, Cassandra, Hadoop, HBase, etc.)’ — though the number of participants was quite limited.

This week’s survey question is about your index size. How big is it? How many documents does it include? What does the rate of change look like?

Don’t forget, this question is your 3d of 4 chances to enter to win an iPad! Winner of the iPad will be announced October 13.

Crawling my company’s ‘intranet’ for HTML docs

NoSQL-style data store (CloudDB, Memcached, Cassandra, Hadoop, HBase, etc.)

No single type of data source accounts for more than 50% of my documents