Ideally, we would have a sizable data set, similar to Rubin's, in flat files with which we would experiment how to update or add files and different approaches.
@patricialarsen, @fjaviersanchez mentioned a couple of experiences and references using simulations, maybe can add them here
Apparently, GCRCatalogs would allow this functionality, but then it would assume that users of the value added catalog would have to interact with the data in the same way.
Ideally, we would have a sizable data set, similar to Rubin's, in flat files with which we would experiment how to update or add files and different approaches.
@patricialarsen, @fjaviersanchez mentioned a couple of experiences and references using simulations, maybe can add them here