This project makes it possible to play with the tools supported by the DigiPres Toolbox, running it in the cloud so you can access it via your browser!
There's a more detailed introduction to the Sandbox on the DigiPres Workbench.
The Sandbox works because it takes advantage of Binder, which was built to support reproducibility in scientific research. By clicking the button below, you can fire up a remote Linux session with a Jupyter interface, and start experimenting...
NOTE please be patient, as it might take a minute or two to before things get started. In particular, the first build of a new version can be very slow and might take a couple of tries. Note also that Binder is an an ephemeral service and your session will be shut down after a while, especially if it's not doing anything.
Alternatively, you can run the Sandbox on GitHub Codespaces, which is a metered service but has a free quota for personal accounts:
Once the Codespace is running, you can get started by clicking here to open the Welcome notebook.
Some test files are included (taken from the Open Preservation Foundation Format Corpus). You can also upload your own files using the Jupyter interface.
NOTE that while any files you upload to this cloud-hosted service should remain private, this cannot be guaranteed. Do not upload sensitive material!
To work on the files in this repository you need Docker and Python 3. You can then install the Binder tools:
pip install jupyter-repo2docker
And then use the ./build-and-run-here.sh script which runs this command:
jupyter-repo2docker -v .:/home/${USER} .
This means the local folder is passed through and you can keep any changes you make to the files.
The layout of the workspace, e.g. tabs and notebook arrangements, can be saved and restored as per the example given here: https://github.com/ian-r-rose/binder-workspace-demo/