[Moose-dev] Re: Distribution Map

I just had a quick look at the article, because I couldn't remember how the order of elements was computed. Now I see we have to port some algorithms from Hapax :) What is the deal between VW Moose and Hapax by the way? Both are released under BSD License, but is it ok to integrate back some stuff from Hapax into PhaMoose (following licence terms of course)? Second question: the article claims that the mapping color-properties is done in such a way that cross-cutting props are shown in distinct colors (i.e. light colors) than well-encapsulated colors. How is this done exactly? Using the focus measure to order all properties and map them from dark to light colors? Is the ordering of elements inside a container computed (apart from the fixed global order): for example, are the most-spread elements more likely to appear at the beginning (because they will appear in many container)? On 20 avr. 09, at 15:38, Alexandre Bergel wrote: -- Simon