KWTR: Semantic desktop

[edit] Contributors:

Knud Möller, Smile Group, DERI, National University of Ireland, Galway

Still under construction...

Please add your CV in the list of contributors

[edit] State of the Art

The term "Semantic Desktop" (SD) generally speaking describes the idea of using Semantic Web (SW) technology in a desktop environment, as opposed to on the Web. On the Web it is online resources which are enhanced with semantic metadata, while in a SD scenario it is desktop objects like files, contacts, calendar entries, etc. that are linked and annotated. Similarly, it is Semantic Web Services and Agents that operate on data on the SW, while it is desktop applications that operate on data in the context of SD.

[edit] History - The Memex and Hypertext

The roots of the Semantic Desktop research field go back a long way. A very early, but also one of the most important influences in the field is Vannevar Bush's article "As we may Think" from 1945 [1], in which he describes a system he calls the "Memex". The Memex allows the user to view a vast amount of (e.g. research) literature that is stored in it, but more importantly, it allows them to create "trails" between related documents while they read them. In this way, a network of documents is created, very much like an early vision of Hypertext. The Memex was never actually implemented, because - until today - the technology was not available. Ted Nelson's Xanadu project (e.g. [3]) and Doug Engelbart's early work on Hypertext [2] extended Bush's ideas. Those three - Bush, Nelson and Engelbart - are often cited as the earliest influences and forerunners of the Semantic Desktop.

[edit] Semantic File Systems

Another important influence in the SD field is the idea of semantic or metadata-enabled file systems, i.e. file systems in which files are organized and can be found by their metadata, instead of (or in addition to) their location in a hierarchical system. Instead of hierarchically organized folders, folders can be defined as queries over file metadata. E.g. there could be a folder with all papers by Norwegian authors a folder with all papers concerning Object Oriented Programming. The SIMULA paper by Dahl et al. would be in both folders. An early and important reference for semantic file systems is a paper by Gifford et al. [4]. Some of those ideas are implemented in modern file systems, e.g. the Extended Attribute (xattr) feature that many Unix and Linux distributions use. In many instances, the Gifford paper also reads like an early design document to Apple's Spotlight metadata file index.

[edit] Current Research

The increasing stability and popularity of the Semantic Web technology as well as the increase in computing power and speed has led to a situation where a lot of the ideas that had been around since the 40's can now be implemented. A number of research prototypes have been developed around the idea of a Semantic Desktop. The Gnowsis [5] system is a Java-based platform that allows linking of desktop data according to user-defined Pimos (Personal Information Model Structures), and stores the generated metadata in a central RDF store. Additional metadata is extracted from desktop objects through Aperture adapters. Haystack [6] is both a SW data browser and “universal information client”, that is extensible through plugins, but does not interact naturally with other applications. Similarly, IRIS [7] is an application framework to allow the development of Semantic Desktop applications, and comes with a number of knowledge management applications, such as an email client, office applications or a calendar. DeepaMetha [8] is a knowledge management platform that organizes, visualizes and gives access to desktop objects based on the Topic Map paradigm.

[edit] Existing Tools and Demos

[edit] Complete Semantic Desktop Systems

• Gnowsis (see above)
• Haystack (see above)
• IRIS (see above)
• DeepaMetha (see above)

[edit] Metadata Indeces

• Spotlight - Apple's metadata file index
• Beagle - a metadata file index from the Linux world
• Strigi - another metadata index from the Linux world, operating on streams instead of files

[edit] Various other Tools

• Kante - A small, light-weight tool to link objects on the desktop. Based on Spotlight.
• Knoten - A small, light-weight tool to integrate Web resources such as RDF file with your desktop. Based on Spotlight.
• Shift - A small, light-weight tool to generate representations of desktop objects (contacts, events, ...) in RDFa, so that they can become part of the Semantic Web.
• SemperWiki - A desktop-based, open-source semantic personal Wiki.

[edit] What are the open problems in your Semantic Web research field?

Research and development on the Semantic Desktop faces a lot of the same problems that the Semantic Web at large faces. The infamous URI crisis is always looming: how do I know the URI for a given thing, how do I know what exactly is meant by a URI, how do I figure out that two URIs mean the same thing, etc.

Apart from those problems, the Semantic Desktop also faces a number of specific problems. One of those is the Integration with familiar working environments: While it is possible to build a complete Semantic Desktop with all desired applications from scratch, this is o

[edit] Relevant Semantic Desktop Research Projects

• NEPOMUK - NEPOMUK is an EU Framework 6 project (FP6-027705) which brings together a significant and large number of research institution and companies to "develop a comprehensive solution for extending the personal desktop into a collaboration environment which supports both the personal information management and the sharing and exchange across social and organizational relations". The goal of NEPOMUK is to build a standard, comprising a software architecture, ontologies, etc. for the Semantic Desktop. NEPOMUK already had a strong influence on the Linux KDE desktop community, where a number of suggestions have been picked up and will be included in the next major release of KDE.

• What challenges try these projects to overcome?

....

• What are their foreseen benefits (both in market and scientific community)?

....

[edit] When will the projects’ results be ready for industry?

I'm confident that, after NEPOMUK finishes in December 2008, the framework that the project establishes will be stable and consistent. The architecture including interactions between all components, inter-application and inter-desktop communication, ontologies, etc. will all be specfied and implemented in a number of reference implementations. At this stage, the framework will be stable enough to be taken over by industry.

• Do you think that it is important to invest (money and time) in these topics? Why?

....

• What are, in your opinion, the most relevant Semantic Web challenges that will be solved in the long term (10 years)? Why?

....

[edit] References

[1] V. Bush. As we may Think. The Atlantic Monthly, July 1945.

[2] D. C. Engelbart. Augmenting human intellect: A conceptual framework. Summary report, Stanford Research Institute (SRI), Menlo Park, CA, USA, October 1962.

[3] T. H. Nelson. A file structure for the complex, the changing, and the indeterminate. In ACM 20th National Conference Proceedings, pages 84–100, Cleveland, Ohio, 1965.

[4] D. K. Gifford, P. Jouvelot, M. A. Sheldon, and J. James W. O’Toole. Semantic File Systems. In 13th ACM Symposium on Operating Systems Principles, October 1991.

[5] L. Sauermann. The Gnowsis — Using Semantic Web Technologies to build a Semantic Desktop. Master’s thesis, Technische Universität Wien, December 2003.

[6] D. Quan, D. Huynh, and D. R. Karger. Haystack: a Platform for Authoring End User Semantic Web Applications. In Second International Semantic Web Conference (ISWC2003), Proceedings, 2003.

[7] A. Cheyer, J. Park, and R. Giuli. IRIS: Integrate. Relate. Infer. Share. In 1st Workshop on the Semantic Desktop at ISWC2005, Galway, Ireland, November 2005.

[8] J. Richter, M. Völkel, and H. Haller. Deepamehta - a semantic desktop (poster). In 1st Workshop on the Semantic Desktop at ISWC2005, Galway, Ireland, November 2005.

[9] E. Oren. An overview of information management and knowledge work studies: Lessons for the semantic desktop. In The 2nd Workshop on the Semantic Desktop, at ISWC2006, Athens, GA, USA, Nov. 2006.