tomay / rebioma

Automatically exported from code.google.com/p/rebioma
0 stars 0 forks source link

taxonomic revalidation #374

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago
Modify existing taxonomic validation system to "revalidate" existing occurrence 
records when a site administrator uploads a new or modified Taxonomic Authority 
(TA). 

Revalidation depends in part on the "validity" of existing records:
If "valid" == FALSE, then check again against new authority. If record fails 
again, no need to re-notify owner. 
If "valid" == TRUE, then check if accepted species, or any higher taxonomy has 
changed (but how will you know this? By checking verbatim species against new 
list?), and update as necessary.

In general there should be no reason to "invalidate" an already "valid" record. 
This may require additional thought and discussion, however. For example, there 
may be cases where there is no direct mapping from old to new, for example 
where an old name has been split into new names based on geography. A few years 
ago Solitary vireo (Vireo solitarius) was split into three species, Plumbeous 
(V. plumbeus), Cassins (V. cassinii) and Blue-headed (V. solitarius), each 
occupies a different summer range in the US. If we originally had a bunch of 
Solitary vireo records, then updated the TA, there is no automatic way of 
knowing which of the new name options to assign the old record to, except 
possibly by examining each location one by one (and even this may be 
problematic on range edges). This case may be one where invalidation makes 
sense, and requires an expert to discuss with the data owner, and assign the 
record to the appropriate new name (by renaming and re-uploading). This example 
also exposes an additional issue - the need to avoid simply revalidating all 
existing Vireo solitarius records because they match one of the new names 
(Vireo solitarius). Many of them are no longer a match for that species.  

There may be additional examples like this. 

Original issue reported on code.google.com by tom.alln...@gmail.com on 5 Feb 2011 at 1:05

GoogleCodeExporter commented 9 years ago
Issue 369 has been merged into this issue.

Original comment by tom.alln...@gmail.com on 18 Feb 2011 at 12:38

GoogleCodeExporter commented 9 years ago
Some additional notes:
- Is there any other validation besides taxonomic validation (e.g. geographic, 
etc.)? Should this also apply here? 
- Be conservative on taxonomic changes - don't create work for no reason, 
ideally synonyms match and names change throughout,  seamlessly to data users 
and reviewers
- Trigger invalidation, and thus (potentially) re-review only when absolutely 
necessary

Original comment by tom.alln...@gmail.com on 18 Feb 2011 at 12:41

GoogleCodeExporter commented 9 years ago
fixed and deployed

Original comment by tom.alln...@gmail.com on 29 Oct 2012 at 11:41