instead of basing it on internal dataset name. this makes it easier to map onto cached datasets without having to rename and do funny tricks like we do on drakh
actually, this is only an issue on drakh because we datalad installed it manually. it should always match the dataset, and in future all preprocs will be named the official dataset name
instead of basing it on internal dataset name. this makes it easier to map onto cached datasets without having to rename and do funny tricks like we do on
drakh