|Agents on the Semantic Web – a roadmap to the future
An aerial view from 10 000 feet
This is the second aerial view, 10 000 feet, is a closer view of the technologies, which are expected to be the cornerstones in the next generation Web. It is the second step in the working methodology to reveal the roadmap to future of the Semantic Web. A visualisation of my approach to find the best predictions is shown in Figure 1.
Figure 1 Working method of the roadmap for Next Generation (the up-side down pyramid)
The main objective of this second step is to better identify the status and trends of research and innovation in the areas of Semantic Web, SWSs and Agents. Google and MINDSWAP, a lab at University of Maryland for demos of semantic Web applications based on DAML “standards” such as OWL. Google has already a range of application products within Search, Explore and innovate, Communicate, show & share, Go mobile and Make your computer work better.
Background knowledge and literature
In addition to the literature described in the “20K aerial view”, is the content of this document based on following work and sources:
Projects at University of Maryland – MINDSWAP
Projects at DERI, Austria
Work on FIPA standards after transition to IEEE
Evolution of JADE
In the layered Semantic Web model, the development is in the phase of creating ontologies for a range of different application domains usually without any harmonizing. The ontologies are typical small and the tools are still requiring some expertise in the theory of ontology. In addition, it is far to many languages of ontology competing to be the preferred standard.
Two of the most distinguished ontology languages at the moment are OWL and WSML. To get a rough estimate of the popularity of the two, a plain Google search with respectively “.owl” and “.wsml”, give the results of 73 700 000 for OWL and only 103 000 for WSML. This should indicate the dominant role of OWL regarding maturity as well as available tools. The oldest and most known OWL toolkit is Protégé. It was developed by Stanford Medical Informatics and is an extensible, platform-independent environment for creating and editing ontologies and knowledge bases.
A range of projects at MINDSWAP show what can come of Semantic Web applications. They already have or are about to develop several demos such as:
To watch what is going on in the field of Web applications, you have to examine new products from Google. The products may roughly be gathered in following groups:
Search products, i.e. a range of products helping to organize and find information about different things, e.g. pictures and videos.
Explore and Innovate products, e.g. find and download open source code
Communicate, show & share products, e.g. share your thoughts and ideas online with a blog that is fast, easy, and free, or calendar to organize your schedule and share it with friends
Go mobile products, e.g. view maps and get directions on your phone; or use text messaging for quick info
Products for making your computer work better, e.g. a Web accelerator that speed up the web on your computer
As a very successful commercial company, Google keeps its cards close to its breast. In general, Google has a split personality when it comes to questions about its back-end systems. To the media, it doesn't want to talk about its technological infrastructure. On the other hand, the Google engineers open the doors a bit when addressing computer science audiences. As a result, there exist sources that describe the Google technology to some degree of details. See a paper written by David F. Carr “How Google Works, July 6 2006”, and “ Google Guide – making searching even easier” by Nancy Blachman and Jerry Peek.
In the next chapters, I will try to compare the trends in the Google technology with the emerging technology of Semantic Web even if the knowledge of the Google technology is quite limited. Especially, I will look closer of how applications within the domain of “Image Search” may be realized in Google and respectively Semantic Web.