uima-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Adam Lally" <ala...@alum.rpi.edu>
Subject Re: Cas Viewer discussion
Date Thu, 07 Aug 2008 20:24:20 GMT
On Thu, Aug 7, 2008 at 12:05 PM, Tong Fin <tong.fin@gmail.com> wrote:
> The approach taken by the CAS Viewer is not new. The GUI design of the CAS
> Viewer is based on "multiple-page form editor" like plug-in/manifest editor.
> In CAS Viewer, each page corresponds to each CAS view (sofa). In each page
> (or CAS view), I use "sash form", "master/details block", and "sections" to
> represent/separate the logical structure of the CAS and to show the
> relationship and interaction between parts. For example, when user selects
> some "type" in the "CAS Types" section, the annotations of that selected
> type will be highlighted in the "Document" section.
> These sections are very related with each other. I don't see much benefit
> from moving from "section" to "view". If we have "multiple-view"
> (multiple-sofa) in the CAS, the situation becomes more complicated.

Okay, from a real-estate perspective, I guess there is not that much
difference between the two approaches.  I think the main point of
having a perspective with multiple views or editors is exactly to
support integration of different components (views and editors).  That
seems to be exactly the situation that we're in now with the CAS
Viewer and CAS Editor.  Wouldn't it be easier to mesh the different
parts of the Viewer and Editor if they were separate views/editors?

> We need to do the integration step-by-step, prioritize the works, and try to
> avoid "over-design".

Well, I think we're stuck at step 1, which is to decide whether to
build a monolithic editor or a collection of different editors/views.
How would you move forward without making a decision?


View raw message