IMPORTANT: please read

Sandy Meier smeier at rz.uni-potsdam.de
Fri Oct 1 23:44:36 UTC 1999


Hi Friends!

Maybe you have followed the controversial discussion on our mailinglist
about the new KDevelop design. This design will be developed by an external
team or as Ralf suggested by serveral teams containing KDevelop developers, the
XXL team and some others. This new version will replace the "offical" KDevelop
in the future and will contain "support for plugins, scripting, different
project types, different languages, new design based on Corba/KOM/OP". and so
on.

We all know that the current KDevelop design is fixed to support C/C++ only
and adding support for other languages is very difficult. (dialog-editor,
classparser, debugger, codegenerator and so on) This is not certainly a result
of a bad design (ok,maybe :-)), it's result of our project goals in the past:
Providing a C/C++ only IDE for easy Qt/KDE development. With the current
approach it's possible to add plugins (C++),add MDI and Dockwidget and an
integrated debugger (which already exist, thanks to John :-)) And there are
plans for a new project managment, you know. 
I think we are at a point where we must decide: Either we start from scratch
with help from the XXL/Kobol team or we continue our work on the current
KDevelop codebase. 

I would like if we continue coding on the current CVS HEAD branch and develop
KDevelop  to a really nice C/C++ IDE with MDI,debugger and so on. Unfortunaly
this means codesplit by the XXL Team.:-( but in my opinion there are only a few
advantages and too many disadvantages for the "start from scratch" approach.

There are some facts which brought me to my actual opinion:
-CORBA/OP will be largely dropped by the KDE Team (see kde-core-devel,cannossa
thread), only some "small" parts of KDE2 will use CORBA
- there are already nice Java/Pythons IDE's : CodeGuide,Netbeans,IDLE and so
on, so more language support isn't much needed 
-KodeKnight (based on KOM/OP design) is dead. :-(, no commits since 4 months

and personal I don't want add features which were already implemented and
fixing bugs that were fixed (current KDevelop). I think this isn't much fun.


Discussion is open!!:-)

Ciao!

Sandy





More information about the KDevelop-devel mailing list