Interface and structure
Film is understood in retrospect, so scenes must be presented before the viewer is asked to make a choice. Knowing this, it is easy to see why the wedding film was perceived as so much better.
The main problem with the travel film interface, is that it is difficult for the user to understand the choices before it is too late. Users can alter the length of the film at any point by clicking a pyramid marker, and they will be taken to a corresponding point in the longer edit of the film. Let's say sequence "A" is over, which the viewer realizes as a different sequence "B" begins. It is only now that the viewer can know that he or she actually would liked to view a longer version of sequence A. But if she or he changes now, it is the longer version of sequence B that will appear.
more...
Film is a retrospective and temporal art form: only after the montage can we know what it is about; only after the sequence can we know how long and detailed it was. It is only after the sequence that we know if we would like a longer sequence, and then it is either too late, or we need to view a lot of material over again.
As the travel film had many unpredictable episodes and catalogs, it was necessary to show parts of everything in the longer film. Still, it was hard for the users to know whether they wanted to change to a longer version or not. As catalogues and episodes accumulate their meaning from the succession of shots, it takes a while before the viewer can form an opinion of what it is about, and it is only when it is over that the user knows how long it was. This means that in most cases, it is when the episode is over that users understand that they want to see more of it, and then it is to late.
Possible solutions to this problem could be to include a timeline-control, like in Hyper-Hitchcock or Kon-Tiki II, or a chapter menu, much like in this hypertext. In the context of my experiment, this was less desirable, as it would be difficult to create on the intended platform of DVD. I also tried to make the stretchfilm as close to regular home movies as possible, keeping in mind a typical viewing situation of several people together in the living room.
Clickable timelines and chapter links in video are becoming quite normal on the web, however, so such options should definitely be tested in future experiments.
(collapse)
The footnote approach of the wedding film is easy to understand, as it presents a taste of a sequence before it gives the user the choice of what to see. Its interface is also more convenient, giving the user time to react, by showing a placeholder (or, rather, timeholder) image or sequence when the link is available.
more...
The footnote approach also has a few practical advantages: In the multiple-versions approach, he same film footage has to be edited two, three or even four times. A technology that adds work will certainly not be adopted by amateur users, and not too eagerly in professional organizations either.
Not only does one need to cut the sequence again, deciding which shots to include and which to leave out. In addition, the different length makes different cutting rhythms natural. When the same scene is presented in a shorter edit, a faster cutting rhythm often manifests itself. That means that even the segments that are included in all versions will need to be trimmed in each instance to fit with the rhythm.
The viewers reacted to the very hectic pace in the shortest version of the travel film, where many episodes were shown in brief excerpts. The footnote structure of the wedding film made it possible to maintain a more relaxed rhythm between the footnote links as well as in the footnotes themselves.
(collapse)
Thus, it is not just the visual interface or the thematic material that made users prefer the wedding film format, but also the editing style. Editing for stretchfilm requires a very straightforward style, where similar shots are grouped together in scenes or catalogs. The reason for this is that it must be possible to represent all parts in the shorter film. This is more effectively done by metonymy, that is, that a tight edit or a short shot represents a larger heap of similar shots.
The two experiments used a special kind of material: amateur video. However, the principles of stretchfilm can be usefully applied to other professional genres, such as broadcast news and fiction film.