Semantic Web – a 'humanizer' for computer-aided work.
20.05.2008
INTERVIEW: Martin Hochmeister aims to create a new discipline of „human-software architecture“ within the next generation web. The expert talks about future developments in an interview led by Marion Fugléwicz-Bren.
Q: What are going to be your personal tasks and priorities within the SWC?
Martin Hochmeister: At SWC, I am in charge of connecting potential customers with the idea of semantically enabled services and products. Furthermore the goal is to generate semantic use cases that will help people to enhance knowledge productivity in their particular work environment. My very personal priority is to prepare the semantic web philosophy in a way that even people who are not too familiar with information technology may get a feeling why semantic web technologies help us structuring information on the web and in our personal environment.
Q: What is your personal attitude towards the semantic web?
Martin Hochmeister: I think that the semantic movement is an opportunity to finally return to a more human kind of computer-aided work. The semantic web connects things with a meaningful relation. Out of this an information network emerges which puts things in a contextual structure. This is much likely the same way a human mind is organising information and experience. It does not mean that the semantic web is a new attempt of realising the AI (Artificial Intelligence) philosophy, definitely not. But from my perspective it provides tools which let us handle information in a more convenient manner than has been the case until now.
Q: How can semantic technologies improve skills management systems?
Martin Hochmeister: Let’s put it this way: Skills management systems in a company’s environment support the human resource department in order to have the right people in the right time at the right place. However, finding the most appropriate employee for a given task is a remarkable challenge. This is because in many cases the skills are not stored on a sufficiently up-to-date level. Executives and employees often update and agree on the skills status once or twice a year during the staff appraisal. But the change of technologies is used to have much shorter cycles than half a year. This is the point where semantic technologies may help. With a tagging mechanism, employees can assess their colleagues using skill tags like “java” or “drupal”.
"A semantic layer can improve a skills management system, if, for instance, some employees use different tags words, but actually talk about the same skill."
Moreover, it is possible to rate those skills by the means of experience gathered through communal project work i.e. after a project’s lessons learned session. A further step is to build an ontology of skills which manages potential problems if some employees use different tag words, but actually talk about the same skill. Such an approach meets the challenge of today’s dynamic technology business and opens room for many more applications in the field of human resources.
Q: Trust is a key requirement in the contexts of the Semantic Web, Social Web and in all kinds of teamwork. What are the issues we facing?
Martin Hochmeister: In the semantic web, context trust has several forms. On the one hand, we have the goal of establishing a reliable, encrypted connection between parties, so that people can be sure that nobody is eavesdropping on them, looking for specific information. This is vital for developing people's trust in the transfer medium. On the other hand we have the aspect of authenticity. Is the partner or resource I am interacting with really the resource I believe it is? Those two aspects are more technology focussed. Probably these challenges will be solved sooner or later.
"Social semantic applications - require a culture of trust and openness, in particular the will to share information for a collective benefit."
Considering the situation from a social perspective, we have the good old issue of trust between humans, which represents the real challenge to overcome. This is valid for both the open and closed environments (companies). The point is that successful social semantic applications - like the skills management system mentioned earlier - require a culture of trust and openness, which means in particular the will of sharing information with others for a collective benefit. I think so far that most of the people are not ready yet to accept those “unusual” values.
Q: How will semantic technologies influence the way we communicate?
Martin Hochmeister: Semantic technologies generate context within an environment of resources. These resources can be persons, documents, topics, whatever. The consequence is that e.g. a person gets in contact with another person because of the same interests. The semantic network enriches the personal focus and reveals new information between people..
Q: How can the potential of the web of relationships be realised?
Martin Hochmeister: Using the semantic web philosophy in a global environment requires the semantic lifting (i.e. the process of enriching with metadata) of data stored in different ways, like HTML or in relational databases. In an effort to accomplish this mission, the Linked Open Data Initiative is one of the first projects which triplifies open content stored in platforms such as Wikipedia, Flickr or GeoNames. So far there are more than two billions of RDF-triples available. The vision is to have a semantic database layer over the web which will finally consist of 100 percent RDF-structured data.
"There are more than 2 billion of RDF-triples at the moment: The vision is to have a semantic layer of 100% RDF-structured data."
It is plain to see that this evolutionary step of the web is coming, since major companies like Google, Microsoft or Yahoo start dealing with semantic web standards as well. A good example for a field that will benefit from this development is information search. People google around a lot and hope to find information by entering half of the answer they are actually looking for. The semantic search will help us here, since it does not only know about the meaning of items, but also about the connections between those items. The classical keyword search obviously has reached its current limit. Now it is time to gradually structure the data in an open semantic format, so we can start to efficiently find the information we are looking for.
Q: Nova Spivack and Semantic Social Software: What is your impression of Nova’s work?
Martin Hochmeister: Nova’s brand new product Twine promises an online desktop onto which we put our documents, contacts, etc.. Twine helps then to organize, share and discover information around our interests. By the means of semantic technologies, Twine automatically organizes information, learns about interests and makes recommendations. The more we use Twine, the better it gets to know us and the more useful it becomes. To be honest, that sounds a little bit scary to me. Nevertheless, I’m sure that Twine is a well considered framework of semantic technologies, at least Nova Spivack is known for decent developments. I got my beta invitation recently and am looking forward to exploring Nova’s work in more detail.
About Martin Hochmeister
Martin Hochmeister started to work as a software designer in telecommunications at Kapsch Carrier Com in Vienna. During his work at Kapsch he moved for almost a year to Ottawa/Canada, where he led a project designated for the European market. He was responsible as a product manager for the development of interactive voice response systems and contributed as a software architect to the development of next generation intelligent network services. Since 2006 Martin Hochmeister has been working as a freelancer, consulting several companies in the information technology business with a focus on information management issues.