I’ve got 15 years of mostly unorganized photos which I imported in PS and now would like to organize a bit more (group by events, either by tagging or by moving them into directories, tagging people, removing bad shots, …).
Since PS does not (yet) have this functionality, I’ll use another tool for that, and I’d like to know some of the consequences.
As far as I understand it, I could do my organization both in the one of the source folders or in the imported folder. Are there any clear recommendations?
I thought of using ACDsee for my changes, as it supports face detection. It can be configured to store the tags in the pictures itself, not as sidecar files. If I PS then syncs the folders again, does it copy the whole file or just updates the metadata?
Note that PhotoStructure does not move assets already in your library (even if automatic organization is enabled), precisely to gracefully handle when people have already organized their photos into events or other albums whose paths aren’t strictly y/y-MM-dd (or whatever pattern you set in the assetSubdirectoryDatestampFormat setting).
The recommendation is to do whatever works for you. My personal library lives on my newer NAS on newer disks, and the originals came from boxes of old hard drives that I’ll never spin up again (probably), so I only make changes to the folders in my library.
Note that you can delete files, move them to new directories, rename any folder in any directory (as long as it’s not in the .photostructure directory), and the next PhotoStructure sync should figure out what happened.
Sure: feel free to use whatever works for you and your library. Know that face detection is the next-highest-voted feature, so as soon as search
If the original file is modified, the SHA will be different, and PhotoStructure will treat the file as a new variant of the prior asset and copy it into your library (if it isn’t already in your library).
If you can configure ACDsee to write to sidecars, then only the sidecar will get copied into your library. The new sidecar contents will be layered on top of prior-extracted metadata.
I personally recommend using sidecars whenever possible, given how easy it is for software bugs to cause issues with your precious original files.
Unfortunately, ACDsee only supports sidecar files for files not supporting metadata directly (the alternative. So maybe I’ll using some other tool instead of ACDsee.
BTW: Are you interested in some example files containing ACDsee-face-detection data?
I know, but I’ve some free time at my hands now, so I’d like to start now
Sure, that’d be great. If they follow industry standards, PhotoStructure should just work, but if they’re adding their tags someplace else, we can add that as a feature request to work on later.
