On 21/02/2008, Phil Steitz <[EMAIL PROTECTED]> wrote: > The 1.2 release VOTE is about to close, but during the vote these two > issues have been opened. > > I have looked at both of these issues and am personally comfortable > with marking them as either for 2.0 or WONTFIX (most likely for > MATH-195), but I don't want to do this preemptively without giving > others the opportunity to weigh in. If others think these need to be > addressed in 1.2, I will stop the VOTE and we can address them; > otherwise I would prefer to proceed with the release when the VOTE > closes, after marking the issues for 2.0. > > I think the fixes in MATH-193 are valid, but not worth holding the > release for. While I appreciate the feedback in MATH-193, I can't > endorse any of the recommended changes, either because I disagree with > them or for backward compatability issues. Other opinions welcome. >
Although the Javadoc fixes in MATH-193 make the text agree with the parameter names, I think that's not the best solution. Short parameter names like "d" are easy to type in code, but they don't work well for IDEs that provide completion hints, e.g. when calling the methods from other code. For example, rather than: public RealMatrixImpl(double[][] d) it would be clearer to use something more descriptive, e.g.: public RealMatrixImpl(double[][] sourcedata) I would suggest deferring the changes. > Phil > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
