I've been working on a project that has morphed into something bigger than expected. It's a continuation of ideas stewing for years. Technology has evolved where some of these things can (and are) coming to fruition.
The project at hand is building a robust system for managing hundreds of thousands of images, PDFs, docs, videos, and other media types for a library/museum. The storage, retrieval, and archiving needs are very specific--have to consider decades and centuries, not just months and years. WordPpress fits prominently into the solution package as does other apps.
WordPress makes for a very flexible back-end development tool, but since it's not designed that way, there are some challenges. My suggestion is to look at the possibility of evolving a version of WordPress in that direction. Done right, it would be useful in a number of industries.
I could give hundreds of examples where this would be useful, and will share with anyone interesting in joining this muse. A muse is all this is at the moment. There are plugins that accomplish some of what's needed. But looking at the big picture, it will require core rewriting (user management, administration, media organization, database storage, backup, and redundancy).
As I was writing all the needs down last night, I keep coming to the conclusion that the solutions needed couldn't be done efficiently (and meet the project requirements) without hacking the WordPress core. Plugins I've incorporated work great, but they don't take the long-(very long)-term needs into consideration.
Plus, there's a need for WP to play nice with other solutions written in other languages and built using different logic schemes (Python, Java, Rails).
My solution is to have--essentially--two WP installs. One is entirely behind the scenes and does all the heavy lifting. The other just serves the pages to the public and (will) have a pretty UI.
My thought: A version of WordPress entirely dedicated to back-end workflow and content management. It could be used in a number of industries and workflow schemes. An added plus is that the public-facing front end could be anything...not limited to WordPress. (similar to what moma is doing http://wordpress.org/extend/plugins/json-api/).
Thoughts? Possible? Worth doing? Yes, it wouldn't have mass appeal, but I can see it being used in newsrooms, libraries, museums, academia, anywhere where back-end workflow is (in many ways) more important that the public-facing website).