Last week I bought myself a sony e-book reader (the touch edition). Sometimes I just love reading fiction (especially when it's snowing outside like now), but I hate buying these books which are then just wasting space in my shelf. The books in my shelf are supposed to be either references or I read them several times. But many books are just like movies: you rent them, watch them, and forget them. Well, so I bought this reader, and I am thouroughly impressed. Double click on any word, and its definition is displayed. It supports highlighting and basic writing functionality (which is enough for e.g. correcting texts). It's a bit difficult to read complex documents, e.g. PDFs based on Word documents with lots of tables and diagrams. Well, it's not difficult, it's simply not possible. But for everything else, it is wonderful.

Our OGC Discussion Paper about Semantic Annotations in Standards is finally online, took us long enough for the final edits. But we are already planning the next version. We discuss how to insert links to external vocabularies or other documentation into existing OGC-compliant metadata (and even data like KML). We realized soon that the theoretical discussion requires some practical implementation. All our implementations, ranging from semantic data integration to semantic validation of service compositions, assume that the Web services or data sets are "already" semantically annotated. We didn't really care where the semantic annotations come from (we of course thought about user interfaces and so on, but not so much about the implications for existing implementations).

Our new (open source) project sapience has been initiated to address this issue. From the website (I am too lazy to write it all over again):

"The Semantic Annotations API (sapience) comprises libraries giving application developers a simple and fast way to extend their applications with semantic functionality. Existing applications with complex and data models usually lack ways to describe the meaning of the data, which unnecessarily impairs the exchange of data across different applications. This is especially the case for Web services serving arbitrary content which have to be integrated into other applications. Due to our background in Geoinformatics, the libraries have a strong focus on supporting the annotation of geospatial content compliant to standards  published by the Open Geospatial Consortium (OGC). The various libraries have been either developed within scientific research projects, or are results of thesis implementations (ranging from BSC to PhD). The implementations have been refactored and simplified for sapience."