On 3/31/11 4:50 PM, Eddie Epstein wrote:
Hi Jörn,
Seems like the idea of visualizing different Sofa types is a complicating
issue. To me it also seems like overkill. Built in basic support for text is
clearly needed. Perhaps just making visualization extensible would keep
the core editor simple. For example, hitting a visualize button for an
audio annotation would invoke some pre-specified plugin that would
play the segment, or open it in an audio editor.
I agree, we should only focus on text, but it should be possible to reuse
as much as possible of the Cas Editor when someone needs to make
a new type of editor to visualize a sofa in a different way.
This could be quite useful for text applications as well.
Consider machine language applications where the translated text in one
view has references to the source text in another view; imagine popping
up a custom window showing both source and target text with
connections between word subsets from one to the other that are
driven by features in the annotation or in referenced types.
That could just be put into an editor which extends the existing
Cas Editor. It would implement the editor differently, but might want
to reuse other existing views, or support to load and store a CAS.
Jörn
