Among other things, this might make scoring and ensembling not as straight-forward.
Another concern is that it is important to be able to sort the bin names into an order but the order may not be alphanumeric. If the bin names are spread out within separate required and optional lists then it is not easy to know how to sort them.
Per https://github.com/orgs/hubverse-org/discussions/24, the thinking here is that it does not make sense for a hub to allow submissions where some cdf output_type_id fields are optional.
Among other things, this might make scoring and ensembling not as straight-forward.
Another concern is that it is important to be able to sort the bin names into an order but the order may not be alphanumeric. If the bin names are spread out within separate required and optional lists then it is not easy to know how to sort them.