Why can't we have our cake and eat it too? I'd suggest that all of these methods be given maximally understandable names... such as sparse_random_normal for sprandn. Can't you then simply define sprandn as an alias for sparse_random_normal? Would there be maybe a to have only the long names exported normally, and then setting up the aliases if you want them...
Scott On Monday, April 27, 2015 at 3:48:47 PM UTC-4, Patrick O'Leary wrote: > > On Monday, April 27, 2015 at 2:36:40 PM UTC-5, François Fayard wrote: >> >> Ok thanks. I did not think about normal. And my background is mathematics >> (and I don't want to know Matlab ;-) ). Imagine how puzzling it could be >> for many people. >> >> It totally violates the "Style Guide" which claims: "conciseness is >> valued, but avoid abbreviation (indexin() rather than indxin()) as it >> becomes difficult to remember whether and how particular words are >> abbreviated." >> > > The chronology goes more like, "we had a method named `sprandn`, then > someone wrote a style guide which contradicts it." As you correctly > guessed/looked up this name is from the MATLAB branch of the Julia family > tree. > > I suspect sparse matrix users will grab pitchforks if you force them to > prepend the full word "sparse" to every one of their method names. (I live > in a dense matrix universe and have no dog in that fight.) >
