At the Max Planck Institute for Informatics, we have been working for several years on producing large-scale knowledge bases from Web-based data. This includes the YAGO knowledge base that integrated Wikipedia with an upper-level ontology based on WordNet. Additionally, we have been investigating approaches to harvest knowledge from unstructured text and integrate it into an existing knowledge base. ResearchCyc would be useful for research on interlinking and integrating heterogeneous knowledge sources. Apart from that, it could also serve as a ground truth dataset in evaluating information extraction techniques.