Proposed inclusion of .b grad file format into specification

[Lestropie]

• Would belong in BIDS raw specification rather than BIDS DWI derivatives, since it is equally applicable to BIDS Raw data

• Would necessitate description within BIDS raw specification of difference between .bvec and .b, as it is not only a transpose but also a change in reference axes between image and scanner space

(Related: Reminder to self to check whether or not the neurological assumption of .bvec was incorporated into the main specification)

[arokem]

See also https://github.com/bids-standard/bids-specification/issues/349

[francopestilli]

see also https://github.com/bids-standard/bids-specification/commit/1d50d14bed67e97892d29ee4edcfbc68a8388d59#r35716642

[oesteban]

Hi @Lestropie, as Ariel and Franco are posting, there is an initiative ongoing that is very much related - your feedback there is very welcome. I guess the most straightforward link is https://github.com/bids-standard/bids-specification/pull/352 (Ariel linked the issue that prompted that PR and Franco linked one of the ramifications of the PR).

That PR was advancing fast, but got stuck with this https://github.com/bids-standard/bids-specification/pull/352#issuecomment-879877873 - I guess the feedback of the three of you would be very clarifying. The person proposing changes actually drafted how they could be implemented in https://github.com/mattcieslak/bids-specification/pull/2

Do you want to take a look and maybe help push that one into the spec including your ideas for the .b grad file format if it cannot be represented as a .tsv file?

[Lestropie]

I suspect I may have missed a number of BIDS discussions where I was tagged because my email has been forwarding anything BIDS-related to a dedicated folder, and that will happen regardless of whether or not I'm tagged. I'll have to try to sort that out so that I actually see anything I'm tagged in.