Stefan Bodewig wrote:
On Wed, 05 Nov 2003, Steve Loughran <[EMAIL PROTECTED]> wrote:

Stefan Bodewig wrote:

Oh, Latin is my third language (and I never had the opportunity to
tell girls I was interested in what they had to say in Latin 8-).

I believe there are still some parts of switzerland and a bit of the Dolomites near Cortina that still speak a derivative of latin, if you feel like trying your luck.


now, they dont have to care what I say, but we do have the option of
pulling all .NET support from ant1.6...


I don't think there is such a significant user base for them, is
there?  You and me, but most .NET developers who are not doing
heterogeneous builds will pick NAnt today (or VS.NET).

I probably would too, and when vs.net shipped would use that, simply because integration with the IDE and code you are working on makes sense. I encountered a fair bit of resistance in getting everyone on the team here to install a modern JRE so they could run Ant to build the C++ stuff. Some people still have issues with CppUnit too.


and that is going to be one of the hard things to change with VS.net. Its very-excellent debugger does bias people towards debug-based development, rather than writing the tests and tracking down problems by writing new tests. In a way, the limitations of the Java IDEs -no dominant tool, historically weak debuggers (better now), made us use Ant and JUnit because there was no way to get things working otherwise. .NET does have a single IDE with good debugging, so people have been able to struggle along with a non-scalable, non-automated process. Or as I call, it "edit-and-continue" programming.

If I ever get the time, I plan to use the .NET tasks with Axis to do better client side interop testing. This is why the <wsdlToDotnet> task is there. I may need <nunit> and <al> shortly too.

-steve


--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]



Reply via email to