Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.stsci.edu/ops/tof/tof_minutes/07-22-99_brainstorming Дата изменения: Fri Jul 23 17:40:38 1999 Дата индексирования: Sat Mar 1 16:34:33 2014 Кодировка: Поисковые слова: http www.badastronomy.com phpbb index.php

Brain storming sessions from 7/22/99 TOF meeting
------------------------------------------------

3) These are the comments that I thought to write down during
the discussion of Andy's write up:

We need to include a disclaimer about how the process is geared
towards getting input from users.

We need to add examples of the types of problems these features are
intended to solve.

Emphasize graphical view by putting it first, not last.

Need to address performance.

Add that "we will strive to make the system as intuitive as possible
by" (using look-and-feel of general purpose and astronomical
software).

Want to emphasize that we will focus on things RPS2 can't do first.
Then move on to replace things RPS2 doesn't do well (or does slowly).

How can we even suggest that we will deliver bright object checking
and guide star info via the VTT when we don't have control over NGSS
development?

Need to establish early on the "Guiding Principles" in the general
vision section. This would be a good place to discuss things like
"capture the why" instead of relegating it to the "other
possibilities" heap at the end. And for that matter let's rename
that section "promising avenues for study".

4) We are eliminating RPS2 and this will be viewed as a good thing.

Anuradha pointed out that we should studiously avoid making VTT and
other new tools backwardly compatible with RPS2. We need to wean the
users away from RPS2 during the transition period. Our strategy should
be to come up with ways to get users to leave RPS2 behind.

Rob suggested that this might cause some users not to try out the new
tools and then we wouldn't get as much feedback.

What are we going to do with RPS2 in the mean time? Will it be
supported so that everything can still be done with it (but maybe
not as easily)?

Will the manager types be uncomfortable that we are throwing away
RPS2? Maybe some, but probably not Steve.

Words Anuradha wanted me to write down:

If you are going to stay up to date with technology you
sometimes have to be willing to throw away code that is
behind the technology curve. The architecture envisioned
in this paper supports this philosophy.