Do it yourself!

We propose you several activities to experiment with advene functionalities.

Annotate the video

You are provided with a schema Noises which contains three text/plain annotation types: laughs, applauses, otherNoises, and some views dedicated to these types:

• a laughsTour STBV that jumps from laughs to laughs
• a report on noises reportNoises in the video, grouped by categories

The first half of the stream has been annotated using this schema, so you can experience the views with some material.

You should now complete the annotation for the second half of the video using the defined annotation types.

• During the movie play, hit the Enter key to begin an annotation, enter key to end it (first select the annotation type you wish to use)
• The import transcription view (put timecode marks on the text of your annotations)
• The annotation edit view
• The transmutation facility in the timeline to use annotations of other types as "temporal templates" : select one annotation, drag and drop it on the annotation type you want to use, and modify the contents of the new annotation.

You can now enjoy the laughsTour and reportNoises views on the base of a complete analysis.

Design a new static view

Write your own HTML report on the Ted Nelson speech at HT'03, using texts of your own, text from annotations content, images extracted from the video, links to play the stream, etc. with TALES expressions (best thing to do is to have a look at the existing views definitions and copy and adapt them).

Design a schema and annotate

Define a new schema called discourse containing at least the joke annotation type. Define a jokeLaughs relation type between joke and laughs annotation types. For this, use the tree view, and the contextual menu.

Annotate the stream with jokes, use the timeline view to set up relations between jokes and laughs (drag from the source annotation and drop on the destination annotation).

Design new dynamic views

Prepare a dynamic view bestOf that contains rules so as to make a "best-of" of the speech jokes. You could also use the utterance annotation to do so.

Design a dynamic view that presents the "success" of a joke, depending on the length of the annotation laughs it is related with (eg. "good" if length >= 2 seconds, "not so good" instead).

A different method for abstracting the speech

As explained earlier, the virtual montage for the D_asbtract view could have been done by designing a summaryCuttingRelation in the speech schema, setting up relations between utterances annotations, and designing a summaryCutting dynamic view that uses the relations to automatically jump in the video.

Give it a try.

Import schema and view, use them for another video

You could reuse the schemas/views defined in nelson.xml package to annotate a new video: Ted Nelson keynote speech at HT'01 conference (DivX File).

Go back to index, to packageConception.