This talk will present the synergy arising from building platforms on top of which do our research. RMOD our team [1] is developing two platforms: Pharo (a dynamic reflective object-oriented language supporting live programming) and Moose (an open-source software analysis platform [2]). Developing platforms forces us to develop really usable systems. While some activities are more engineering than research per se, it is really interesting to deeply understand problems or impacts of certain design decisions. Developing platforms is rewarding because it is more a long term effort and ensures a degree of stability. Platforms also often exhibit non-linear growth that is really exciting. Finally this setup raises many interesting questions such as “What is the value in terms of citations or published papers of a couple of end-users”, or “Is it not really stupid not to work on latest hype language?” To try to open our minds, I will draw parallels with the notion of wealth of an ecosystem in biology. In the second part of the talk I will present some selected results around Pharo and Moose such as: automatic minimal system core generation, dynamic core updates, selector namespace, dependencies in past commit branches and automatic migration rule generation.

[1] http://rmod.lille.inria.fr
[2] http://www.moosetechnology.org/