In this talk we present the experience of the software produced, as part of the research activities, at the French Gaspard-Monge Computer Science laboratory (Laboratoire d'informatique Gaspard-Monge or LIGM in French). The LIGM has an important production of research software with 66 software items identified for the period 2013-18, where 50 have been disseminated as FOSS. Since 2006, the Lab has worked in order to improve the dissemination and the visibility conditions of this research production [1], which includes the adoption of policies and the production and maintenance of a catalogue. In this context, one of the first issues we had to deal with was the concept of the Lab's software, in order to answer the questions of the researchers. The second concept was the identification or reference (or citation form) of such objects. The concept of "Logiciel d'un laboratoire" [1,2] initially proposed in 2006 has since then evolved to become research software ("Logiciel de la recherche") and it is a central concept in order to propose services and infrastructures to deal with this production in the current Open Science global context evolution [3,4], where the goals are to make it visible, accessible, reusable. It is also a central concept in order to propose research evaluation procedures related to this scientific output [5].