Living at the base of Cape Cod. Enjoying the ocean life.
In my workflow tags == keywords which are orthogonal to albums. Keywords describe a photo and get published with the photo. One photo can be in multiple albums like Flickr Photosteam (published to Flickr), “Zion Vacation” (event album, published to Piwigo for friends and family consumption), “2022 Year in Photos” (yearly highlight album), “Favorite Landscapes”. Photos are in a date hierchary on disk.
DigiKam has come a long way, but it’s still missing some really basic DAM features. I’d love to stop using Lightroom and there are plenty of good raw converter replacements, but it’s the DAM features that there are no replacement for. In DigiKam an album is a directory on disk. A photo can’t be in two albums at once. They recommend using tags for more groupings, but that just seems clunky.
Create an automation to map the central scene events to actions. In triggers, choose the device, then the choose a trigger something like “Central Scene action on endpoint 0 Scene 001”, and value something like “keyPressed 2x”.
I use automation like this to make double press and triple press on one switch activate multiple devices.
Of the two, /i would go with D3. I loved D2 when it came out and tried going back with D2 Resurrected, but it wasn’t the same. D3 is a fun game, but after playing D4 in beta and release, I don’t see me ever going back. POE 2 maybe.
Unfortunately, some of those users are bringing the worst of Reddit with them. Their is an explosion of low value cutesy comments.
I’d be fine with tags in the digiKam DB including all of the above, but the values written to the images IPTC keywords field would have to be limited to the subkeys of what you listed above as type and content hierarchies. Here’s an example of one of my images: https://www.flickr.com/photos/tcgoetz/53265234720 you can see what I use for keywords in the info below the image.
My workflow also includes grouping similar images, using compare to grade them best to worst, marking the best with the pick flag, and rejecting the worst. I also group subimages of panoramics with the generated panoramic as the top of the stack and the pick image. didKam has a group feature but it seems different. Grouping in LR hides members of the group other than the pick image.