charvolant
commented
1 year ago Clarification requested. Can you give an example, please so that we can trace the sequence of requests.
In the biocache, a query like https://biocache.ala.org.au/occurrences/search?q=taxa%3A%22Cracticus+tibicen%22 has the following sequence:
- Cracticus tibicen is looked up in the name matching service via https://namematching-ws.ala.org.au/api/search?q=Cracticus%20tibicen This will return the accepted name and the left- and right-values
- The biocache-ws does a SOLR search for records with a left-value between 40591 and 40607 (subject to change with a new name index)
- The resulting list is formatted and returned
Via the BIE, the sequence would be
- https://bie.ala.org.au/search?q=Cracticus+tibicen
- The first link is to Cracticus tibicen (Latham, 1801) (accepted name: Gymnorhina tibicen)
- Clicking on 'View Records' on https://bie.ala.org.au/species/https://biodiversity.org.au/afd/taxa/5291343e-fdeb-4a65-8ba5-928f5b96acf5 goes to https://biocache.ala.org.au/occurrences/search?q=lsid:https://biodiversity.org.au/afd/taxa/5291343e-fdeb-4a65-8ba5-928f5b96acf5
- The lsid is looked up in namematching-ws and the search proceeds as above.
Sherrin-ALA
type 1 errors technically aren't errors “ it's actually a limitation of using BIE that users should be clear about. IE “ BIE will always return the exact answer to the query name it is given. Therefore it will always return a subset of the total number of records if there are more than one possible taxonomic combination and therefore it is an unreliable means of querying the data for number of records at any taxonomic level. The only reliable means of querying the data and getting a correct answer is to use the front page, navigate to the species page and then use the classification page to check for actual or ALA-caused synonyms that may have additional data attached in the taxonomic tree.