Establish "NMIND publication" checklist?

[yarikoptic]

As a part of the BIDS Steering group we were approaching different publishers about possibly adding some indication (e.g. banner) for the papers which are accompanied/process/etc BIDS compliant datasets. In the discussion we decided to may be just expand such an effort where BIDS use/compliance would be just one of the items. But then we are also getting into the realm of COBIDAS checklist, https://github.com/ohbm/eCOBIDAS (ping @shnizzedy @Remi-Gau et al) is the app. I am yet to research on how much of overlap with COBIDAS here but would be interested to hear from NMIND and eCOBIDAS how much of synergistic opportunity there is and how we could ensure that BIDS use/compliance is "verified" during publications review phase and then evidently made visible on the publications across publishers.

[shnizzedy]

I'm in favor of adding a BIDS-requirement item that's N/A-able (in case the tool is more general, like datalad), something like

• requires BIDS file naming

maybe in bronze infrastructure?

Or we could maybe make it non-N/A-able like

• supports BIDS file naming

?

I lean towards requiring BIDS where applicable, at least for outputs.

Re: COBIDAS, I think there's a lot less scope overlap there than with OpenSSF Best Practices
; even though the "M" in NMIND stands for (once stood for?) "method", the checklist here is tool-focused and COBIDAS is actually methods focused. To me, that one seems complementary to this one.

[Remi-Gau]

Couple of things.

intended targets

COBIDAS is way more for the typical neuroimaging study where you acquire "brain data" from a bunch of participants and do something with it (or not in the case of a data paper).

NMIND checklist seem more targeted for typical tools for neuroimaging (pipelines or not).

So the overlap won't be null but I expect it to be small. I definitely think that some items from the NMIND checklist could / should make their way into COBIDAS. The opposite seems less true.

BIDS

BIDS and cobidas

When people publish a paper and their data is said to be accessible I think that it should be mentioned it they are BIDS valid and reviewers should actually check this. This is not part of COBIDAS. Yet.

BIDS and NMIND

It should be mentioned if a tool can ingest a bids dataset or is BIDS aware in some ways. This a requirement for BIDS apps for example.

If that is the case then it should also be mentioned if the output will be BIDS friendly (i.e follow the typical BIDS filename pattern of [key-label_]*suffix.ext). This is unfortunately not a requirement for BIDS apps.