It would be good in the next iteration to include everything that's in the MPC in the simulated database or pick a random subset of simulation objects that match roughly the discovery numbers today and set discoverysubmissiondate to NULL to mimic having things that are being detected that are a previous before-LSST discovery. I imagine especially early in the survey for comets and candidate ISOs, we'll want to know if this in a new object or something previously known and the discoverysubmissiondate is an efficient way to do that, so it would be good to test code using this.
It would be good in the next iteration to include everything that's in the MPC in the simulated database or pick a random subset of simulation objects that match roughly the discovery numbers today and set discoverysubmissiondate to NULL to mimic having things that are being detected that are a previous before-LSST discovery. I imagine especially early in the survey for comets and candidate ISOs, we'll want to know if this in a new object or something previously known and the discoverysubmissiondate is an efficient way to do that, so it would be good to test code using this.