Hey all,
Some of view pointed invalid BIDs datasets on openneuro.
I have opened issues to report those and added comments on those
datasets. If you find more datasets with egregious problems, don't
hesitate to report them similarly or to pass the info to me.
Another problem that can sometimes happen is data curators
"forcing" their datasets to be valid by telling the validator to
ignore certain files. This can lead to certain "bidsignore
abuses". We opened an issue on the openneuro repository to
centralize those problematic datasets so we can best see how to
address this problem.
https://github.com/OpenNeuroOrg/openneuro/issues/2419
On a related note, I will try to see if there are ways that bids-matlab can be used for certain aspects of quality control that go beyond pure bids-validation (i.e making sure that all subjects have same number of files / events / conditions, detecting "empty values" in tsv...)
@+
Rémi Gau