Advocating for the basic sciences has always been an important, and often complex task.  Lately, it seems like it is not enough to share results with the public at large, but to get them involved in the methods.  The Afterglow Project will connect people from all over the world and teach them about the current theories, but also share the state of the art in modeling methods.

To that end, The Afterglow Project will provide access to interactive and informative videos, podcasts, applets, and community spaces to facilitate learning and comprehension.  Astrophysics is a very visual science and there are a lot of places within afterglow physics where interactivity can be integrated directly into the output of certain models.  Additionally, we are striving to have a high level of interactivity with the distributive computing project based on the original boxfit routine.  We want people involved and active in every step of the process, from data ingest to finalizing the parameters.

All code powering the Afterglow Project is open source.  The original boxfit routine can be found on the NYU afterglow page.  It was built and is maintained by H.J. van Eerten.  The scientific paper introducing it can be found here.

We are currently expanding on this codebase to create something that is GPU accelerated.  First on CUDA, and the on OpenCL.  This will allow anyone with a sufficiently modern computational setup to participate in the project.  There are currently some port hurdles coming from meshing the memory management schemes of the original program with the limits imposed by current GPUs.  This code will be available on the project github as soon as it is stable.  A complete rundown of the issues and progress can be found on the development blog.