rosemaryjoconnor
commented
1 year ago Meeting today with Arthur. He will be sending sample exports for the 2 databases together with sample R scripts with database connections.
There is no API on their end but the databases are available for public access. Images are available and can be placed on a server for access. Directories of images are named with directory name matching sampleId, so there is some structure there to build image URLs
There is no real IT support capacity for the project. Arthur was hoping for monthly updates eventually, so likely a fetcher will be the go. Looking at the database the structure is not complex and any additional information we need he is happy to add.
From Prof Arthur Georges:
We have a couple of databases (one for lizards in our breeding colony, but including a lot of wild-caught animals; one for freshwater turtles with about 36,000 samples). They are hosted on AWS in MariaDB. Both are living databases in that they are in continual use with new data entered continually. Almost all of the wild caught specimens have lat longs associated with them.
The idea would be to have some sort of automated system that mines our databases for specimen records with locality data etc and uploads those that have not already been uploaded to ALA via some sort of API (provided their Status is 'public' and their locality data and taxonomic identity has been verified).
I already have the database for Pogona linked to a suite of scripts in R, so it would be a matter of moving the data from R to ALA. I can handle the scripting at this end, if someone can handle it at the ALA end.