Tweaked my scripts for special handling of Barkerville photo records, since they have different mapping to images, and also have Chinese caption data we can use. In addition, I made the bash script output curl each of the images from their website, which worked a treat; I now have 578 new records in the db, and I'm waiting for HBR to run the CSV import for 485 images matching them. I think that's all the low-hanging fruit from Barkerville ingested now; time to see if we can get a raw db from any of the other repos.