Ryan Ly

A new `Headstage`, `Amplifier`, and `DataAcquisitionSystem` types, which all inherit from Device, would also be useful. They need to relate to one another such that it is clear how signals...

I am reviving this because it came up in a discussion today. Not urgent but I am recording my notes. We should use ProbeInterface to the extent possible: https://github.com/SpikeInterface/probeinterface They...

> This has always been annoying. Maybe we should just create a separate data type for external videos I am also in favor of this solution and would prefer it....

> > he new data type `ExternalImageSeries` (?) could inherit from a base Events > > I'm fine with creating a separate type, but I don't fully understand why this...

After discussion with @bendichter, we decided to: 1. Add a new type `ExternalImageSeries` which inherits from `NWBDataInterface`. It contains a dataset of external file paths and an attribute indicating the...

@bendichter How are labs using storing time series of multi-channel image data in the NWB file, if at all? According to the `dims` and `doc` for `ImageSeries`, the data of...

See also https://github.com/NeurodataWithoutBorders/nwb-schema/issues/316 where we changed `OpticalSeries` to have dimensions (frame, x, y, rgb)

Also worth noting with this change is that all types that extend ImageSeries and use external files will need to have a corresponding new External type. E.g. `ExternalTwoPhotonSeries`. That's doable...

To summarize this issue and PR: Problem: The `ImageSeries` neurodata_type is a subtype of `TimeSeries`. `TimeSeries` requires the dataset 'data'; however, `ImageSeries` makes 'data' optional because 'external_file' may be provided...

After discussion with @oruebel, we have decided to go with Option 1 for now and put the other changes in this PR regarding handling of external files on hold for...