One of the restricted keywords for EEG coordinate systems is BESA:

https://bids-specification.readthedocs.io/en/stable/99-appendices/08-coordinate-systems.html#eeg-specific-coordinate-systems

The way this keyword is currently described is ambiguous, and it seems to suggest that the BESA keyword refers to an idealized, spherical coordinate system instead of a measured coordinate system (which will in practice never be on a sphere).

The link further describing BESA (see here) is furthermore ambiguously referring to three different coordinate systems:

• "device" (this coordinate system is irrelevant for us)
• "head" (this coordinate system is identical with another keyword we have: Captrak)
• "besa" (the spherical model we are also referring to in the spec text)

This situation should be resolved for the following reasons:

1. BIDS is about data, not copying templates --> we want to share measured electrode positions, not positions that are computed (simulated) on a sphere. Such position templates can be supplied by the analysis software.
2. Even assuming the case that BESA was unambiguously referring to the "head" coordinate system, we'd have a duplicate with the Captrak keyword

This error was brought to @robertoostenveld and my attention in a recent talk with @kucint and brain products.

For a fix, it'd be great to just delete the BESA keyword. Unfortunately, that change is not backward compatible.

The alternative is to write clarifying text and add it to the spec (and remove all ambiguity).

Dear EEG and BIDS people, hat are your opinions on this situation and potential solutions?

Furthermore, the clarifications for Captrak should be extended as well, currently we are only referring to that information on the fieldtrip site.