klíčová slova
a  b  c  d  e  f  g  h  i  j  k  l  m  n  o  p  q  r  s  t  u  v  w  x  y  z 

Ars Electronica Online Archive and the Visualization of Semantic Networks

Technical aspects of Ars Electronica Online Archive (software, administration), as well as details about the ongoing project Visualization of semantic networks.

Ars Electronica Online archive and Visualisation of Semantic Networks

Dietmar Offenhuber: Right now, there are lots of media art archives, but they are all limited project, maybe because they got funding and then stopped operating. There are a lot of data distributors around the world but it is all very fragmentary. One would have to combine and view all of it in order to make sense of it. This is one part of the situation, the other is that in books about new media art you always encounter the same set of artists, the same superstars that are in every publication, and the rest is completely invisible. One of the goals of the visualization research group is to provide this context of what is happening, what is actually stored in these archives, and how it relates to material that is already published. We try to develop tools to explore all this massive amount of data and find trends and influences.
Lucia Udvardyova: Are you developing some kind of software?
DO: We are developing software. Although because of limited resources, we only focus on few basic aspects. We have narrowed our focus on three areas of research, one thing is dealing with networks – our archival system that we are programming right now stores the data as a semantic network because we thought that is the most flexible way to combine all these different fragments that are around since they map to the same ontology (what is an author, project, to make sense of the structure of the archive). We try to develop new ways of how to interact with these networks because the usual way is that you type in few constraints then you wait and get image as a result, and if you are not happy with it and want to change something, you have to repeat the whole process. We have already developed tools to help sculpt the data.
LU: Is that the tool that you have developed in-house?
DO: Yes.
LU: What platform does it work on?
DO: This is an older project. I developed it together with a collaborator from scratch. It’s a network browser SemaSpace. We also want to rely as much as possible on existing tools and frameworks.
LU: Could you specify which frameworks?
DO: It depends on the purpose. For example if you want to put something online so that everyone can experience it with the least barrier, you typically go with either Java script or Flash what pertains visualization interfaces for the archives because it’s most accessible for people. But if you want to provide a research tool that people would download, then for example, Java is not so great in web environment.
LU: Will you make the software you have been developing accessible for other archives as well?
DO: Yes, certainly. That is the whole point of it. Everything is open-source, it only makes sense if it is used. It is always difficult because every institution has different demands.
LU: Will this work on all types of audiovisual archives?
DO: Other as well. One of our main archival frameworks is Fedora, which is widely used. This is our basic platform, then we add facility for the semantic network data structure and then we developed the visualization tools that come on top of it.
LU: What kind of technology is the Ars Electronica Online archive built on?
Gabriele Blome: TLP protocol.
DO: This is one of the three. We also have different approaches, information landscapes. Here the networks are not in the centre but rather quantifiable dimensions of the data (how similar, how dissimilar they are, with this information, build a mental map of the archive). This is a different tool that we are building, and the third approach will deal with text, to support online reading of text and also combine full text with semantic network.
LU: What would you say that are the specifics of audiovisual archives?
DO: One of the specifics of AV archives would be to provide annotation of the timeline of an AV piece. One of our collaborators already has a prototype that supports it.
LU: When will this be available?
DO: The first release is scheduled for 2009 but it is not certain which part of the project will be completely implemented by that time.

The interview was conducted as part of the project supported by the Ministry of Education, Youth and Sport in Czech Republic, NPVII-E (human resources) entitled "Metodika výuky a formy vědecké spolupráce s použitím audiovizuálních archívů v mezioborovém prostředí".