@haakon@aus.social @xssfox@cloudisland.nz @saera@ausglam.space
thanks for the tag Craig,
There are very few options here because no matter how you publish, the underlying platforms could be traced back to you.
Some options:
1. Create an account on HF or other repo like Zenodo using an email account that is relayed to your address through an anonymiser, like Moz Relay. It's not perfect but it's another layer of abstraction.
2. Upload your dataset
3. Create an account on ArXiV using the same relayed email address, however you will need someone to validate you in the field you're publishing in because your relayed email won't end in .edu*. This person should preferably have some separation from you to preserve anonymity (e.g. if a friend does it, you may be identifiable because of your relationship).
4. Published a pre-print on ArXiV and reference the dataset on HF / Zenodo
This is not perfect but provides some degrees of separation.
There is a broader use case here - but the tradeoff is credibility - if the research is anonymous, how can I trust it?
1.3.6.1.4.1.61513
Craig Askings