Now I have had a chance to relax after the AT effort, I have been thinking
about the timing of the forthcoming release. The conclusion I have drawn is
there is just so much in the current CVS HEAD (both new features and
important bug fixes) that our users will greatly appreciate compared to
PLplot-5.1.0, that it is really important to get this release out as soon
as possible. Thus, I think we should forgo worrying about most of the
minor issues that I have detailed in PROBLEMS. That really just leaves
two issues:
* More cross-platform testing required especially for platforms with tcl/tk.
* Fix the -dev tk problem.
Once those two are addressed I will want to move fast for the reasons above.
So any help you can give on either issue would be most appreciated.
While discussing the situation with Joao, he has come up with an excellent
idea; we have a short pre-release period (after the two issues above are
dealt with since we don't want to embarrass ourselves) where we invite our
users to try everything they can to find bugs in the new plplot. I think we
would get a lot of tests if we widely advertised this pre-release period
(which I assume would be a week), and it is all great publicity for PLplot.
Also, I think there should be a prize of Maurice and Geoffrey taking the guy
who finds the most bugs out to dinner....;-)
Does a short official pre-release period with wide advertising seem
reasonable?
Alan
email: ir...@be...
phone: 250-727-2902 FAX: 250-721-7715
snail-mail:
Dr. Alan W. Irwin
Department of Physics and Astronomy,
University of Victoria, P.O. Box 3055,
Victoria, British Columbia, Canada, V8W 3P6
__________________________
Linux-powered astrophysics
__________________________
|
