Using CERN's SWAN & Zenodo DOI

[afrankenthal]

Hello! I'm planning to do a SWAN-based talk since we need CERN's analytix cluster. We will try to make our CMS repository public but it is hosted at CERN's GitLab instance. I'm not sure if Zenodo DOI would work with that, i.e. if it's not on GitHub? I don't think Binder would work with this setup either but if you have any recommendations that would be appreciated.

Thanks,
Andre

[matthewfeickert]

Hi @afrankenthal.

We will try to make our CMS repository public but it is hosted at CERN's GitLab instance
...
I don't think Binder would work with this setup either but if you have any recommendations that would be appreciated.

The public mybinder resource does support arbitrary Git repo URLs ("Git repository" option in the drop down) so as long as your repository is public to the world then you can just paste the full URL in there.

If you need private CMS internal resources though you'll have to use SWAN.

I'm not sure if Zenodo DOI would work with that, i.e. if it's not on GitHub?

Zenodo's repository importer only works with GitHub as far as I know. Though you could manually upload it to Zenodo.

[afrankenthal]

Thanks for the info @matthewfeickert! I'm thinking to make a standalone version that I can upload to Zenodo, with just a little bit of data included and a local spark master, so that people can test it out if they want. But in the live demo I will use SWAN so I can connect to CERN's HDFS storage and process a more meaningful dataset (which the general public naturally doesn't have access to).

[matthewfeickert]

That sounds great to me! Looking forward to it. :)