Also included is a simple Python module 'ninemlcatalog' for convenient access to the models stored in the catalog using the NineML Python library.
The NineML Catalog is maintained by the NineML committee.
To "install" the XML models in the NineML Catalog simply clone the repository to somewhere sensible on your local computer (e.g. $HOME/git/nineml-catalog), and then you can reference the models from other NineML documents using either relative or absolute URLs. Before cloning, it is best to create a fork of the central repo (https://github.com/INCF/nineml-catalog) so you can backup any modifications to your own GitHub repo, and then open merge requests with the central repo (see Contributing).
To install the python module you will need to install the NineML Python library. Then simply add the 'python' directory in the catalog repository to your PYTHONPATH. Once NineML Python library is installed you will then be able to run the unit-tests by the command:
$ python -m unittest test
from the git repository directory and it will attempt to load and validate every model in the catalog.
Contributions to the catalog are most welcome. To add a model or amend an existing one simply make the changes to your local model, push them to your GitHub fork and open a pull request to the master branch of the INCF fork with a brief explanation of what your model models or amendment fixes (see https://help.github.com/articles/using-pull-requests/).
To make merging with the central repository feasible it is strongly recommeded that you make any distinct sets of changes in separate feature branches from the central repo's master branch and then merge them together to create your "develop" branch of the catalog with all your customisations.
Before opening a pull request, please add the author information and relevant scientific citations to comments within the annotations block of the document. It is also encouraged to create or link your model with an entry on Open Source Brain (see http://www.opensourcebrain.org/docs#Creating_Your_Own_Project).