On Sat, Oct 29, 2011 at 09:10:12AM +0200, Luc Maisonobe wrote: > Le 28/10/2011 23:30, Gilles Sadowski a écrit : > > Hello. > > > > I think that something is not quite right in those class names: > > UnivariateMatrixFunction > > UnivariateRealFunction > > UnivariateVectorialFunction > > > > To be consistent, we should either > > 1. indicate the "grouping" and "number" type everywhere: > > UnivariateRealMatrixFunction > > UnivariateRealFunction > > UnivariateRealVectorialFunction > > 2. assume that the "number" type is "Real", and drop it everywhere: > > UnivariateMatrixFunction > > UnivariateFunction > > UnivariateVectorialFunction > > 3. assume that the "number" type is "Real", drop it everywhere, and > > indicate the "grouping" explicitly everywhere: > > UnivariateMatrixFunction > > UnivariateScalarFunction > > UnivariateVectorialFunction > > > > The same goes for "Multivariate...", "Differentiable...", and the optimizers > > naming. > > > > I'd say that option 2 is the least disruptive, and most practical as > > UnivariateFunction > > is more common (and shorter) than > > UnivariateScalarFunction > > You're right, lets go to a more consistent scheme. Option 2 is fair.
What do you think of Ted's suggestion: Vectorial -> Vector ? Gilles --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org For additional commands, e-mail: dev-h...@commons.apache.org
