The SciViews Box, or svbox, is an easily installable software environment for reproducible data science with R, Python and Julia. RStudio & Jupyter(Lab) are also available.
A Reproducible Software Environment for Data Science
Reproducible research requires reproducible computing environments. There are many ways to standardize software ecosystems: virtualization, containerization (Docker), remastered live linux, virtual environments (Anaconda), cloud computing, … All these solutions require a certain degree of technical skills to deploy. This is a barrier to distribute work to the mass. The easier-to-deploy solutions are thin containers with just a few software components. In case a more complex system is needed, the technical barrier resurfaces.
The SciViews Box is a reproducible software environment for data science providing R, Python, Julia and thousands of pre-installed packages, with RStudio server and Jupyter interconnected in the same box. Installation and use have been polished to be as easy as possible on Linux, Windows, or MacOS. It may be useful to distribute hybrid computations (say R + Python, or R + Julia), or for data science courses (see for instance http://biodatascience-course.sciviews.org/sdd-umons/decouverte-des-outils.html -in French-). In comparison to a course-dedicated server, students truly own their svbox and can use it after the course. This is a major advantage for the adoption of reproducible data science practices by users like biologists or social scientists.
Easily use the svbox with the launcher
The SciViews Box Launcher is a native application on Windows, MacOS or Linux that runs on the host. It starts or stops the SciViews Box running as a virtual machine (VirtualBox). It also gives direct acces to RStudio, Jupyter, JupyterLab, Adminer, or an ssh session running inside the svbox from the host.