This project aims to:
- estimate the carbon cost induced by travel to conferences;
- perform a sequence of analyses related to the geographical distribution of the participants and other patterns of participation to help organizers to take this issue into account.
We applied in particular the tool to the SIGPLAN set of conferences (POPL, PLDI, ICFP, SPLASH) and describes this experiment and its conclusions in an ongoing draft.
The Python script used to perform the analysis is however meant to be as general as possible. We would love to see other groups perform similar analyses. As such, we welcome all feature requests, issues and pull requests. A brief overview of the script can be found in Readme_script. A fairly detailed step by step use of the script can be found in tutorial. More documentation to come.
############### Dependencies ###############
This project uses Python3. It relies on the pycountry, pycontry-convert, request and geopy
libraries, all available through pip.
If you have pipenv installed on your system, you can run pipenv shell and then pipenv sync to get a shell with the
right dependencies installed locally to the directory.
############### Structure ###############
input
Contains the cached db and locations as well as the db of airports. The data to be analyzed must be stored there as well.
output
Contains all computed data to be exploited in folders per analysis ran.
src
Source code of the Python script.
paper
TeX source for the paper describing the analysis of the SIGPLAN conferences.