This page refers to a selection of the software tools I have been involved in, either as a programmer, or as a designer. This list encompasses very different kind of developments: ones that are research prototypes that act as a proof of concept(s), and ones that involve a quite long development processus over several years and that have been used by many users.
This tool, written in Java, is a research prototype [CN.5] that uses an Anoto pen to let a biologist explore a scientific datasheet, i.e data printed on an Anoto-patterned paper, in order to display data properties associated with specific area of the sheet. The tool lets the biologist annotates his/her explorations, and go back to the relevant data and information when pointing on the handwritten notes. (démo à IHM 2009 [in french]). I have developed this tool while working at In Situ. The demonstration at the IHM 2009 conference (in french) won an award (2ème prix) (project information).
biok (Biology Interactive Object Kit), written in XOtcl, is a tool for researchers in biology that enables them to perform bioinformatics analyses and program new ones. I have developped this tool during my PhD thesis, based on participatory design with biologists and academic work on end-user programming, participatory programming [CI.3] and algorithmic flexibility [W.2].
Strip'TIC is a prototype for air traffic controllers that combines augmented reality, digital pen, multi-touch and gestures to connect passive paper strips to a digital system. I have participated to the design of Strip'TIC as one of the designers of the team, through ethnographical analyses, participatory workshops with air traffic controllers, students projects advising, academic work [CI.8] and design sessions (project information). In Strip'TIC, I have contributed to the design of a tangible computation feature, where stack exit time are displayed according to strip placement in a simulated sequence (see figure).
Mobyle [RI.2] is a model-driven software tool, written in Python and XML, that enables to specify Web 2.0 interfaces for bioinformatics analysis programs running in Unix. I have developped the earlier version of Mobyle, Pise [RI.1]. In Mobyle, I have lead the development team for several years and developed prototypes for the Web interface and the distributed workflows [W.7] (see also: Mobyle and Bioworkflow projects). Pise and Mobyle have been used by thousands of users all over the world, and have also been at the origin of many variants and projects (see a review by Don Gilbert). Projects include GPipe, a pipeline generator, an editor (PID and BMID), Inquiry, a platform developed by Bioteam, an Apple spinoff (see this comment [in French]). Pise offered APIs included in the bioperl and biopython open source projects, and an interface for the EMBOSS suite. Mobyle is currently installed on many sites, such as the RPBS platform for structural biology, the INRA bioinformatics platform Migale, the INRIA SERPICO platform, the SNAP workbench at North Carolina State University, etc. Mobyle also offers web services, and is integrated in several related projects.
- [other developments...]
During 8 years in the industry and 15 years at the Institut Pasteur, I have developed numerous tools as a programmer, a database expert or a software engineer. For instance, I have been involved as a programmer in a remote monitoring platform to develop the core event dispatcher, etc.
Domain Company Projects bioinformatics Institut Pasteur Mobyle, biok, databases, BioNetbook,... electronic war Genicorp/Thomson databases portfolio management Genicorp/Paribas databases remote monitoring Genicorp/Cegetel real-time monitoring system staff management Genicorp/COGEMA databases