IBM UFO Repository

Currently, WWW, large enterprises, and desktop users suffer from an inability to efficiently access and manage differently structured data. The same data objects (e.g. Product) stored by different databases, repositories, distributed web storage systems, etc are named, referenced, and combined internally into schemas or data structures differently. This leads to structural mismatch of data that often consists of the same semantic objects (e.g. EBay and Yahoo! online auction offers). Web 2.0 offered a multitude of mashups, microformats, and tagged data without a convenient way to access, deploy, and exchange them. Yahoo! Pipes, Microsoft Popfly, Google mashup editor, IBM DAMIA [22], and other mashup fabrics entered the arena here to bridge the gap between differently structured heterogeneous data. Large enterprises, having suffered for years from the Data Integration Curse , could improve their situations with new data management tools built with significant help from the research community [15, 8, 7, 9, 10, 17, 22, 2, 11, 19, 14, 23, 26, 12, 3, 6, 13, 4, 24, 20, 5]. While this work represents significant progress that surpasses WWW and Web 2.0, none of the solutions are unsupervised yet, and they require significant human effort. Finally, biological and medical domains have many public and proprietary databases storing probably the most complex and large structures (e.g. human anatomy) in the world. All of them suffer from the same curse and are a subject of significant body of ongoing research in BioInformatics (e.g. [17, 18, 21]). In all these domains the cornerstone of the problem is in structural mismatch of data from different sources and insufficient data modularity to make it more accessible. Monolithic data representation (e.g. one large XML file) with abundance of low-level storage/representation details are hard