26th of August - 2nd of September, 2000 Frank Nack EU project meeting (Eisenstein) in Glasgow and NILE WS in Edinburgh Glasgow meeting (28th of August) --------------------------------- Participating partners of this meeting were: GMD-IPSI, CWI, University of Glasgow, and ENST Missing Partners: DFI and BFI The major discussions in Glasgow were concerned with the architecture, copyright issues and the extension of the consortium. Architecture: What we agreed on is: There is a clear distinction between the data (i.e. video, audio, text) and the annotation, there is a clear distinction between representation of knowledge and its presentation. The design of the information network will be completely in the hands of the domain experts. The relevant tools (developed by GMD) will allow them to specify the types and importance of relations, and the types of nodes. How their design is finally transformed into the database, which also means how the database design is reflected by their work, is up to us developers (Glasgow Uni).Regarding the language we decided on XML (knowledge representation in XML Schema). The user interface, e.g. browser and rule editor to control the interface, will be designed by the developers (CWI and ENST). However, we agreed that the publisher should have a saying about the presentational issues and thus will have control over the 'presentation rules'. Hence, the interface for these matters should be easy enough that a designer can work with it. Information about the user (i.e. user profile) is stored on client side The presentation side will be SMIL and MPEG-4 -> BIFS. All sorts of digitization tools will not be included into the system, but the results, i.e. digital video, shall be easily incorporated. Copyright: We discussed that quite a bit but didn't find 'the solution' - it all depends very much on the business model we will follow and it depends on how far we can test the environment during its development. This means, if we test say the user interface in an environment that provides us with access to say 10.000 users - we have to provide some copyright protection. If we use a secure server approach, we can avoid that but will only have a limited platform for case studies. Other options we discussed were: - the filmic material can be provided on a DVD which should be bought - Access only via subscription - There was a suggestion by DFI that we could think of having MOSFILM as a partner (they own most if not all of the Eisenstein films). The situation is that Russia belongs to European countries which can be partner without being funded. However, if it can be duly justified that the particular partner is essential for the achieving the objectives of the project an exception can be made. Extension of consortium: We had the feeling that the business side is still weakly represented in the consortium. I reported on interest by France Telecom and Starlab on becoming partners. Both options were discussed, were Starlab sounded more promising, since they suggested to turn the results of the project into a company which is going to market the tools and as well as the midelware services. The second extension would be related to the inclusion of a publisher (at least the call is directed towards new ways of electronic publishing). Keith suggested ELSEVIER as a partner - not only because he has quite good contacts to them but mainly because they are already investigating new ways of electronic publishing. Keith will investigate what can be achieved here. In general we all felt that a publisher will strengthen our chances substantially. NILE WS (30th of August - 1st of September) -------------------------------------------- The idea of this ws was to discuss the following issues: - What do experienced story tellers do... and what do they want children to learn? - What is know about the similarities and differences between story creation and story telling software and other modes of story telling and story creation? - What is required in design environments to utilise narrative ideas? - What kind of tools are need to support the software design process? - How can the "narrative effectiveness" of learning environments be evaluated? The WS was quite interesting from the point of view that the participants covered a variety of domains, e.g. video/film producers, storytellers, researchers covering all sorts of domains (agents, db, educational computing, animation, etc), www entrepreneurs, etc. The general feel was that story telling is a very powerful way of communicating ideas and it would be feasible to integrate it as a presentational technique. However, it became clear over the two three days, in particular the second day showed that quite obviously, nobody really knows how to do it. There were suggestions for architectures and descriptions of potential systems but none of them were implemented or even close for attempts to do so - though they represented in the best way possible the type of 'design study on an imaginative basis'. Those presentations describing an implemented system focused on environments which support the design and creation of stories, where the actual process was not supervised by the system but rather by a human (e.g. a teacher). Though it was fun to see the videos of happy children using the tools and producing cute kind of stories of all kind of fairy tale personal (prince, witch, frog, etc.)acting in their everyday environment (e.g. school, home, etc.). The stories for teaching abstract concepts, like programming, were all oriented towards the model of a lesson - rather dull. Particular interest caused the aspect of interactivity - most of the time either in a VR environment or on the web in relation with film (broadband applications). In particular the latter seemed to be a problem, because the diegetic power of the media makes it very difficult to combine the material in any way (continuity is the problem here). The displayed examples were really not that interesting (ready designed stories where the user had to solve a problem, e.g. try to identify the killer). The presentation was linear in time, but the created space was divided in different rooms. The user could only be in one place at a time - information gaps were the result. It seems that our instinct regarding the use of narrativity is ok and we do not have to feel irritated because we have not really figured out yet how do do it - nobody really has. For a more detailed investigation on ws topics see : http://cbl.leeds.ac.uk/~paul/nile/index.html The article I submitted can be get from here: ~nackStuff/Articles/2000/NILE.doc