External users (student training)

I think this is not possible, but would like if possible to confirm:
having a CERN computing account is a strict requirement for acessing SWAN, right?
the reason I ask is because I coordinate a summer student program in HEP, where we teach root and analyses, and was looking for an online framework to allow that;
the program is attended by undergraduate students, with no connection to CERN (yet) and thus no CERN account;
some CERN tools become accessible e.g. via a lightweight registration, but I presume this is not the case here; I understand SWAN is used for Open Data analysis, and I wondered whether there were plans to allow some light version available to external users (students) for training purposes.
many thanks and regards, Nuno

1 Like

I’m interested also in this possibility for the same reason. Please let us know if there is a way to do so.
Best Regards
Antonio

Hello,

There is unfortunately no open SWAN instance, and swan.cern.ch requires a CERN account. It’s been a recurrent topic in the past years, and in the end it has not been pursued due to lack of resources - although I agree it would certainly be useful for cases like yours.

That being said, there is ScienceBox:

https://sciencebox.web.cern.ch/sciencebox/

i.e. an installable version of SWAN. You have to manage it yourself, but it gives you freedom to configure your own users.

I think binder might be a good option for you:

Alternatively there is Colab by Google: https://colab.research.google.com/

Hello Artur, all,
sorry to come back to this. I’ve tried to launch the pyroot notebooks on binder, but it doesn’t start since it goes idle and gives and an error. Any idea on how to correctly write a working environment.yml ?
Thanks a lot
Antonio

Hi Antonio, if you’re using conda you have to follow this: Managing environments — conda 4.6.1 documentation
Alternatively, if you just need pip you can do the requirements file: User Guide - pip documentation v21.2.1
I’d recommend using a virtual env like conda or venv to test the files locally before going to binder.

In general, Binder has a category in the jupyter discourse: