After extended work, :BaseKB Now has been productized and is being produced on a regular weekly schedule.

:BaseKB Now is an extract of Freebase data processed by the Infovore framework. Like other large RDF data sets, Freebase contains ill-formed, invalid, and repetitive triples. :BaseKB Now divides Freebase into a stream of rejected facts and 12 streams of accepted facts.

Since many streams state the same information with different vocabulary, and because not all streams are required for all queries, the 12 streams can be combined in any combination to produce the specific data set you need, minimizing processing time, speeding up your application and reducing hardware requirements.

:BaseKB Now applies a rigorous process of Parallel Super Eyeball 3 validation, which ensures that it works correctly with RDF standard tools.

Available releases of :BaseKB Now can be seen on the basekb.com home page, and you can subscribe to the RSS feed to be notified of new releases. Going forward we will retain copies of Freebase source data and publish the exact application code used to produce each release of :BaseKB Now to Maven Central so that our results can be reproduced by following a few simple documented steps.

Weekly releases of :BaseKB Now is published in the Amazon S3 cloud in a requester-paid bucket; access within the us-east-1 zone is free, fast and efficient. A complete download of :BaseKB Now to locations outside of the AMZN cloud costs about $3.

We're planning to release quarterly releases in BitTorrent format with an unlimited retention individual to serve the needs of academic researchers for a product that is stable in time; making it possible for researcher to compare results against identical data sets and to make their results reproducible in the future.

If you're interested in the following the rapid development of :BaseKB Now and the Infovore framework, please star the Infovore Framework on Github, follow it on Freecode or join our discussion group.

Please direct inquiries to ontology2@gmail.com