As a historian working on a dissertation, I'm looking for a good, open-source, Mac-friendly metadata tool for managing the gigabytes of digital photos I've taken of archival materials. Adobe Lightroom's features are close, but it's an expensive piece of software and doesn't solve my problems well.
Mostly, I want a way to add arbitrary metadata to an individual image or batch of images. It might be citation-related ("Every JPG in this folder or its subfolders is from the Jane Texas Papers, and this subfolder is from Box 1..."), subject-related, author-related, or idiosyncratically related to my own research questions. (Think of a sociologist or anthropologist coding interview data.) In an ideal world, it would be able to support a "transcription" field. JPEG/EXIF metadata fields seem unwieldy, not to mention that they're easy to clobber accidentally.
I've looked into DSpace as a repository system (and even installed it on my laptop), but it's overkill for single-researcher use at the dissertation level, not to mention a huge amount of work to set up. I don't have access to an institutional DSpace for building my own collections, or else that might be a good answer (from what I know of it).
Really, what I want is a research workflow system for digital reference images. I want to be able to easily, quickly mark and categorize the items in my (very large) collections according to arbitrary categories; find the items I'm looking for using a faceted search; transcribe them in some cases, but not all; know what I haven't looked at yet, but need to; cite the item I'm looking at; etc. Zotero isn't really suited for this, nor is Bookends (my preferred desktop citation manager).
If such a tool doesn't exist for making this process easier, are there established best practices for setting up a home-hacked metadata and workflow system of my own? I've been hacking around with Ruby scripts that store this metadata in nearby text or YAML files, but those haven't worked as well as I'd like.
