🚀 go-pugleaf

In article <2AE62C23.27686@ics.uci.edu>, echarne@ics.uci.edu (Eli B. Charne) wri
tes:
|>
|>   I'm interested in any pointers people may have to research/papers/etc
|>   that discuss design methods for large software systems to support
|>   independent development and evolution of the user
|>   interface.
|>
|>   It is my experience that the user interface tends to be very tightly
|>   bound to the system that is being designed, and that the user
|>   interface code tends to be mixed in directly with application code,
|>   making it very hard to split up their development.
|>
|>   I would greatly appreciate any response.

   I am working in a research lab (French CNRS - Particle physics
domain) We have developped a toolkit for building interactive
Motif-based application, that make them entirely independent of the GUI
: the UI definition is done entirely interactively during the
application's life (note that the aim of this toolkit is to build
interactive data analysis aaplications) . An online UI builder (linked
within the appli.) and a C interpreter (named CI) makes this possible.
The C interpreter is used to handle any possible callback script
installed in the widgets. The only communication between the
application and it's UI is done through entry points and variables
belonging to the appli. that may be declared to the CI. The important
design point is that the syntax of the entry points (well of their
arguments) is only relevant to the appli. semantic, whereas the 
conversion format from the UI forms is handled by the CI.

   This package contains in addition support for 3D graphics within
dedicated widgets, and support for creating additionnal widgets (data
viewers, imaging widgets, etc...).

   Package + doc may be reached from anonymous ftp on

      lalux3.in2p3.fr

   ---> pub/onx

 Christian Arnault - Guy Barrand

  arnault@lalux1.in2p3.fr
  barrand@lalux3.in2p3.fr

  ----------