urapadmin
commented
7 months ago an "interpretative model" could be a solution, though: - #2528. That's not tampering with the original observations it is putting an alternative model on top of them. That way I could allow to override an original observations within the model but without ever touching the original observation. The original observation could be disabled in favour of an interpretational relation or even simply disabled (instead of deleted).
It will be requested, no doubt. But there should be archaeological reservations (manipulating observations at your desk? Isn't that Pippi Longstocking? Drawing your Harris Matrix the way you like it best?
And then there are technical reservations: This is changing live data live. That has time zone (and hence synchronization) implications and runs into concurrency issues. So I will fight that until we have #293 fixed. And that is nowhere near.