Alex Rousskov wrote:
> Primarily due to faith reasons (i.e., belief in something for which there
> can be no proof), I guess. The amount of work required for an upgrade is
> also an issue, of course. I'd stopped arguing for C++ migration long time
> ago, but reserve the right to bug people about deficiencies of C. :)
I'd say it does not matter much if we select C or C++. Regardless of
language we need to define module boundaries and adhere to these. The
current practice of having lots of deep structure references is not a
very good design, especially not without any form of specification.
I mainly agree with Alex. Using C++ would solve a couple of problems,
both in design and reliability, and with the high availabilty of GCC on
various platforms the porting argument is not that big.. Also the move
to C++ can be done gradually, by cleaning up the design and move part
for part into object oriented design with data encapsulation. But it
alone is does not magically solve our problems, or is a nessecary for
getting them solved.
/Henrik
Received on Tue Jul 29 2003 - 13:15:58 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:12:09 MST