Loss of hard disagreement following taxon swaps

General
1

I’ll start by saying that I’m not a curator and know nothing about the mechanics of taxon swaps or anything - I’ve merely occasionally seen the aftermath as an identifier. I’m also not in any way wanting to criticise those who do such necessary work. My aim is solely to observe that there is sometimes a problem and attempt to brainstorm ways to fix or avoid it. I’d love to come up with a definite feature request or bug report.

This post was sparked by a previous post, https://forum.inaturalist.org/t/original-topic-title-taxon-name-change-dropped-2-ids-failed-to-update-added-disagreements-dropped-during-taxon-name-change/75788, where it was pointed out that a taxon swap had resulted in hard disagreements to a higher level disappearing (and presumably also removed the ‘as good as it can get’ tick?). One example of how it’s worked is https://www.inaturalist.org/observations/332100161 . This (as far as I can tell) is purely the species/complex being moved to a different genus, and has been declared not a bug - which seems odd to me.

I’ve also occasionally observed problematic behaviour in the past with other changes. The one that comes to mind is when Leucopogon was split into Leucopogon and Styphelia: genus IDs were correctly updated to tribe (Styphelieae), but any hard disagreements to genus were lost in the process. This is clearly (I think?) a more complex case, because maintaining the hard disagreement with a tribe ID could result in unintentional disagreement with the correct species if it’s still in Leucopogon.

On a related note (though nothing to do with retaining hard disagreements as such), Boronia and Cyanothamnus are a similar case, except that Boronia genus IDs were not moved up, meaning that they sometimes now disagree with the correct ID they would once have agreed with.

So, with those three cases as examples (and I’m happy to hear more), is there any way we can improve the system to avoid such problems? My feeling is that hard disagreements should always be preserved, but I’m not sure whether that’s possible, since it doesn’t seem to happen? However, along with that - or instead, if it’s not possible - it seems to me that anyone with such a hard disagreement should be being notified if the disagreement is being moved to a higher level, so they can review. Though again, I don’t know what’s possible.

Questions:

  1. If you’ve observed issues like this arising, are there similar-but-different types of cases I’ve missed?
  2. Do you have any concrete suggestions on what could be improved?
  3. Is there any way to identify specific observations that may be problematic in terms of either of the last two examples so that identifiers can be notified and asked to review, either before or after a taxon split? (Preferably with an Identify link to any relevant observations. Am I shooting at the moon?)
  4. (Am I just coming from a place of ignorance to be even trying to ‘fix’ things as complex as this?)
6 Likes
2

There was a recent taxon change with tree frogs that resulted in some issues due to a name overlap (not the technical term) in the system…not sure if people are still having issues or if it worked out.

2 Likes
3

This was/is a huge mess and a great example for issues with loss of hard disagreements during taxon changes. I don’t know enough about how taxon changes work behind the scenes to propose a solution, but this particular change lead to thousands of (previously correct) disagreeing IDs/CIDs being good as can be being wiped out.

Some previous discussion here:

https://forum.inaturalist.org/t/taxon-swap-removes-disagreements/40556

and probably elsewhere on the forum

2 Likes
4

This is a known issue. A bug report was filed but I don’t know if it qualifies as a bug. The problem can occur as a result of the simplest taxon change, such as name change. At least there’s a workaround that can be used to restore the disagreeing IDs after the fact.

1 Like
5

When a taxon is moved from one genus to another, any previous genus level confirmation becomes a disagreement. This is why I stopped agreeing on a higher level.

This problem is different: taxon swaps modified the original ID order to the taxon swap order thus removing the disagreements.
The fix is to retain the original ID order during taxon swaps. One way of implementing this is to retain the original ID datestamp, just add the swap date to the swap comment.

2 Likes