Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.stsci.edu/ops/tof/tof_minutes/05-13-99_brainstorming Дата изменения: Wed May 19 00:37:37 1999 Дата индексирования: Sat Mar 1 16:34:21 2014 Кодировка: Поисковые слова: storm

Brain storming session from 5/13/99 TOF meeting
-----------------------------------------------

First we had a demo from Jeremy of the latest release of the SEA.

* The SEA is a prototype being developed for NGST and we can adopt the
portions of it that we like into our design.

* Currently the interface has a tree representation of the proposal on
the left and a window for the currently open tool on the right.

* But the data and the view of the data are separate in the design and
the user will be able to choose a view to work in.

* We need our design to have an open growth path so that tools (like
the ones we like from the SEA) can be integrated easily one at a time.
We need a pluginable design.

* We need to limit ourselves - not try to create the ultimate tool.
What are the most important tools?

* The latest release of the SEA includes a very rudimentary exposure
planner which will allow the user to specify the relationship between
exposures.

* They are working on new ideas for a graphical exposure modifier.
Maybe a pie chart that shows how the exposures are using up portions
of the 96 minute orbit "pie". Doesn't have to imply a time order, just
proportions.


Next we discussed Nick and Rob's "look and feel" concepts to make sure
we understood the major concepts that they were advocating.

* As with the SEA there is a concept of a workspace.

* There are objects that can be added to the proposal. These objects
are not necessarily completed targets or visits. It is more like a
filing cabinet with generic objects, imported objects, completed
objects, etc.

* Objects may be aggregates of exposures (a three color picture of a
planet) without implying an ordering of the exposures. So one could
say: give me another three color picture only this time the target is
a different planet.

* Nothing in the interface should imply a time order, but there should
be a way to enforce time order if it is appropriate.

* There are tools for operating on the objects. These can work both on
objects and groups of objects.

* There are a variety of views for looking at the objects.

Of these points that three most critical seem to be:

1) Objects for proposal
2) Views of proposal
3) Tools to work on objects or groups of objects


Anuradha mentioned that we should learn from the Gemini project and
not give the user freedom they don't know what to do with. People
like structure. (Some Gemini proposals prepared using their tool
never opened the shutter!)

Both Karla and Roeland expressed the concern that we are putting
emphasis on the environment and not concentrating on the critical
missing functionality. (Like the VTT which many people are trying to do
for themselves.)

But Anuradha and Nick both felt that unless the reader of the white
paper can see a concept for what it would be like to "drive" in this
new environment, they might not be able to understand how the new
functionality fits in. Just a list of concepts is too nebulous.

In our plan we should definitely advocate getting new tools (like some
of the SEA tools) out to the user as fast as possible.

We need a blueprint for bringing together diverse tools.

How can we make the framework generic enough to grow instead of get
old (like most of our current frameworks)?

But we also need to emphasize what this will allow you to do that you
couldn't do before.